Defect Levels Research Articles

ASFormer: Attentive Semantic Feature Fusion Transformer for Pixel -level Defect Detection

Abstract Surface defect detection is pivotal for ensuring product quality in various industries. These defects typically manifest as low background contrast, substantial variations in shape, and a scarcity of balanced training samples. Traditional defect detection methods often encounter limitations in terms of low detection accuracy and insufficient detection granularity. This study introduces an innovative attentive semantic feature fusion transformer network (ASFormer) tailored for pixel-level defect detection tasks to overcome these challenges. Initially, the transformer architecture is employed to extract the multiscale features of defects. Subsequently, a multiscale semantic fusion module is integrated, complemented by a dynamic upsampling mechanism, to mitigate the loss of detailed information during feature fusion. Moreover, a scale-aware dual-attention module is developed, which effectively captures the intricacies of the fused features across both channel and spatial dimensions, ensuring pixel-level detection precision. Additionally, a contextual boundary loss function is proposed to augment the ability of the network to discern defect boundaries, different categories and defect scales. Experimental validation on the NEU-Seg and Crack500 surface defect datasets demonstrated that the proposed ASFormer achieved state-of-the-art performance, with mIoU scores of 85.24% and 78.20%, respectively.

Optical and photoelectric properties of nanostructured SnS films obtained by spraying ink using a nanoparticle suspension

Abstract In this work, SnS films were prepared using spraying ink with a nanoparticle suspension. The average size of the synthesized nanoparticles was (18-20) nm. The structural, optical and photoelectric properties of SnS films were investigated using different characterization techniques. XRD and EDX results show that the investigated films exhibited an orthorhombic SnS phase with a composition close to the stoichiometry (CS/CSn = 0.99) and low level of microdeformation (ε = 1.8×10-3). In addition, the hexagonal SnS2 and tetragonal SnO2 phases were also observed. The presence of SnS and SnS2 phases is confirmed by Raman characteristics. The band gap of the SnS, SnS2, and SnO2 phases was determined using the novel ACFD method based on the analysis of the spectra of the first derivative of the absorption coefficient, which directly determines the energy of both band-to-band optical transitions and transitions involving defect`s levels. These results correlate very well with data obtained using photoconductivity spectra. The nature of the electronic optical transitions as well as the type and energy position of various defect levels were established. It was shown that the energy of direct and indirect band-to-band optical transitions of SnS compound correspond to 1.72 eV and 1.16 eV, respectively. At the same time, the band gap of SnS2 phase equal to 2.05 eV. The ionization energy of the acceptor (233 meV) and donor (100 meV) levels that determine the p- and n-type conductivity of SnS and SnS2 compounds, respectively were defined. Due to its properties, SnS films may be suitable for the development of novel effective solar cells with SnS absorber layers.

Discovery of ART0380, a Potent and Selective ATR Kinase Inhibitor Undergoing Phase 2 Clinical Studies for the Treatment of Advanced or Metastatic Solid Cancers.

One of the hallmarks of cancer is high levels of DNA replication stress and defects in the DNA damage response (DDR) pathways, which are critical for maintaining genomic integrity. Ataxia telangiectasia and Rad3-related protein (ATR) is a key regulator of the DDR machinery and an attractive therapeutic target, with multiple ATR inhibitors holding significant promise in ongoing clinical studies. Herein, we describe the discovery and characterization of ART0380 (6), a potent and selective ATR inhibitor with a compelling in vitro and in vivo pharmacological profile currently undergoing Phase 2 clinical studies in patients with advanced or metastatic solid tumors as monotherapy and in combination with DNA-damaging agents (NCT04657068 and NCT05798611). ART0380 (6) has a favorable human PK profile suitable for both intermittent and continuous once-daily (QD) dosing, characterized by a dose-proportional increase in exposure and low variability.

Development and validation of a nomogram for predicting cerebrospinal fluid leak after endoscopic craniopharyngioma resection.

To identify risk factors for cerebrospinal fluid (CSF) leak after extended endoscopic endonasal surgery for craniopharyngiomas and develop a predictive model for predicting postoperative CSF leak. Six hundred and sixty cases of craniopharyngioma (training cohort: n = 462; validation cohort: n = 198) were retrospectively reviewed between October 2018 and May 2024, and relevant risk factors were identified. A nomogram was built using a stepwise logistic regression method based on the Akaike information criterion. The performance of the nomogram was evaluated using area under the curve (AUC), calibration curve, and decision curve analysis. The overall rate of postoperative CSF leak was 4.5%. Higher prognostic nutritional index (PNI) level (OR 0.819, 95% confidence interval [CI] 0.735-0.912; p < 0.001) and larger dural defect (OR 6.789, 95% CI 3.112-14.807; p < 0.001) were identified as independent predictors for postoperative CSF leak in multivariable logistic regression analysis. The AUCs of the nomogram were 0.870 (95% CI, 0.782-0.957; p < 0.001) and 0.842 (95% CI, 0.722-0.963; p < 0.001) in the training and validation sets, respectively. Calibration curves in the training and validation cohorts showed satisfactory agreement between predictive and actual outcomes (p = 0.608 and p = 0.564, respectively).Decision curve analysis further confirmed the clinical usefulness of the nomogram. Higher PNI levels may help reduce the risk of postoperative CSF leak, while a larger dural defect size was demonstrated as an independent risk factor. We developed and validated a nomogram for predicting CSF leak after endoscopic craniopharyngioma resection, which showed strong predictive performance and could assist clinicians in formulating personalized treatment strategies.

XSCAN: Explainable solder joint defect probability prediction through solder paste printing status with imbalanced data

This research addresses challenges in Surface Mount Technology (SMT) related to solder joint quality prediction, focusing on the initial solder paste printing stage. Recognizing that over 50% of defects originate at the printing stage, this research delves into establishing a direct correlation between printing quality and joint quality. Traditional approaches have limitations in accurately predicting defects due to isolated treatment of printing quality indicators, scarce explainability of prediction models, and lack of joint defect data. This research introduces a novel framework, XSCAN, aimed at predicting the probabilities of solder joint defects from the states of the printed solder paste. This is accomplished by using a generative adversarial network (GAN) to synthesize additional defect data and segment the feature space of printing indicators using customized decision trees to minimize defect probability prediction error. Specifically, XSCAN optimizes generative model structures using decision tree prediction results focused on defects, generating valuable defect information to help feature space partition. Also, pruning rules are designed to handle imbalanced data and improve defect prediction. They enhance explainability by defining safe and high-risk zones for solder paste quality. XSCAN outperforms all other baselines when tested on real-world datasets of chip resistors. It achieves the lowest prediction error and provides different warning levels for potential joint defects. XSCAN takes a proactive approach to improve manufacturing quality while addressing data imbalance and model explainability challenges. It provides practical insights to enhance SMT processes and reduce waste and rework costs.

Defect-engineered WOx/ZnIn2S4 Z-scheme heterojunction boosting photocatalytic H2 production via photothermal coupling

Recent interest in photocatalytic water splitting has intensified the demand in the development of photocatalysts capable of harnessing the full solar-spectrum. This study introduces a novel WOx/ZnIn2S4 Z-scheme heterojunction, prepared by depositing ZnIn2S4 (ZIS) nanosheets onto WOx nanorods, enabling efficient photothermal-coupled photocatalytic H2 evolution. The success relies on the engineered oxygen vacancies within WOx nanorods, which not only confer excellent photothermal properties lowering the reaction barrier but also create defect levels in WOx facilitating Z-scheme electron transfer from these levels to the valence band of ZIS. Consequently, the optimized WOx/ZIS heterojunction exhibits a remarkable H2 evolution rate of 33.91 mmol h−1 g−1 with an apparent quantum efficiency of 23.6% at 400 nm. This study provides a new strategy for developing efficient Z-scheme heterojunctions with broad-spectrum solar hydrogen production capabilities.

Surface double defects-dominated TiO2 with high liquid phase stability for smart SERS sensing of dye additives in foods

Surface defect engineering is an effective strategy to regulate the physical and chemical properties of semiconductors for a wide range of applications. Herein, a surface-highly-reduced anatase TiO2 (R-TiO2) with double defects (surface oxygen vacancy defect and Ti3+ energy level defect) was developed as a plasmon-free surface-enhanced Raman scattering (SERS) substrate for ultrasensitive detection of food additives. Abundant surface oxygen vacancies enable R-TiO2 to exist stably in liquid phase, which realizes a smart SERS detection for dye target molecules due to evading fluorescent interference from analytes, while SERS enhancement of analytes cannot be observed on the solid phase substrate at all. Due to the joint contribution of surface double defects, a multi-channel charge transfer mode and a stronger Herzberg-Teller coupling between substrate and analyte are formed, which greatly enhance SERS effect of the substrate for sensitive detection of dye additives in foods with an enhancement factor of 1.6 × 106.

Low-temperature reduction of NOx with NH3 using a hybrid system of non-thermal plasma-LaMn0.9CO0.1O3 perovskite nanocatalyst

The catalytic reduction of NOx at low temperatures presents a significant challenge, which can potentially be overcome by integrating plasma into the process. Porous perovskites emerge as promising catalysts for plasma-assisted NOx removal. In this study, perovskite catalysts LaMnO3 and LaMn0.9Co0.1O3 were synthesized using the auto-combustion sol-gel method with different fuels and subsequently evaluated for their efficiency in plasma-catalytic NOx removal in the presence of NH3. Analyses including XRD, BET, FESEM, TEM, FTIR, DRS, and TGA indicated that while the samples exhibited similar crystal sizes, they displayed varying levels of defects. Notably, the LaMn0.9Co0.1O3 catalyst synthesized with citric acid as the fuel (designated as LMC-Ci) demonstrated a highly porous and homogenous network structure, characterized by cubic-spherical particles measuring 20–30 nm, as corroborated by the XRD data. The co-substitution of cobalt in LaMnO3 enhanced light absorption, enabling the catalyst to effectively utilize ultraviolet photons produced by the plasma. Among the evaluated catalysts, LMC-Ci achieved the highest activity, resulting in a 98 % removal rate of NOx. The study further examined the influence of operational parameters, including discharge time, temperature, space velocity, and specific input energy, on the performance of the selected catalyst, revealing that plasma presence significantly reduced the operating temperature to 80 °C.

Decisive role of organic fluorophore and surface defect state in the photoluminescence of carbon quantum dots

Carbon quantum dots (CQDs) hold significant promise for applications in biological imaging, sensing, and optoelectronic devices owing to their superior photostability and low toxicity. Nevertheless, the elucidation of their photoluminescence mechanism remains an open question, necessitating further comprehensive investigation. In this Letter, CQDs exhibiting ultraviolet (UV) and white fluorescence were isolated through silica gel column chromatography separation of the crude product obtained from a one-step solvothermal synthesis. CQDs with different luminescent properties exhibit the same crystal structure and similar particle size distributions. Both CQDs exhibit orthorhombic structure where C60/C70 molecules are located at lattice points, having average particle sizes of 2.71 and 2.98 nm, respectively. Consequently, the luminescent properties of the synthesized CQDs are predominantly governed by their surface structure. The results of microstructure characterization and spectroscopic analysis demonstrate that the UV emission originates from the C(=O)OH and C–O–C related luminescent moieties within organic fluorophores, and the blue emission band is attributed to defect states related to surface group C–O–C, while the green/yellow emission arises from C(=O)O related surface defect levels. These observations have gained a profound understanding of the luminescent genesis of CQDs, broadened the luminescence coverage wavelength range of CQDs, and enriched the family of CQDs materials.

Design and analysis of inorganic tandem architecture with synergistically optimized BaSnS3 top and AgTaS3 bottom perovskite Sub-Cells

Perovskite materials are revolutionizing the solar cell (SC) industry, continually enhancing their properties and establishing a prominent photovoltaic technology. Among these, BaSnS3 (BTS) and AgTaS3 (ATS) stand out for their strong potential as absorber layers. These inorganic chalcogenide perovskites address the drawbacks of their organic counterparts, being both lead-free and non-toxic, thereby making them highly suitable for photovoltaic (PV) applications. The exploration of BTS and ATS as absorber layers in a tandem solar cell’s top and bottom cells has yielded remarkable outcomes. The innovative tandem solar cell design features a top cell structured as n-WS2/p-BaSnS3/p+-MoS2 and a bottom cell configured as n-WS2/p-AgTaS3/p+-GeS. This theoretical study using SCAPS-1D demonstrates a high efficiency of 42.57 % with a VOC of 2.03 V, a JSC of 23.29 mA/cm2, and an FF of 89.85 %. These impressive results are achieved with adjusted layer thickness, carrier doping and defect levels, highlighting the strong potential of BaSnS3 and AgTaS3 photoactive materials. The findings reveal the viability of innovative, all-inorganic perovskite-based tandem solar cells, offering a promising avenue for future sustainable and high-efficiency photovoltaic device technologies.

Prolonging Charge Carrier Lifetime via Intraband Defect Levels in S-Scheme Heterojunctions for Artificial Photosynthesis.

S-scheme heterostructure photocatalysts, distinguished by unique charge-transfer pathways and exceptional catalytic redox capabilities, have found widespread applications in addressing challenging chemical processes, including the photocatalytic reduction of CO2 with a high reaction barrier. Nevertheless, the influence of intraband defect levels within S-scheme heterojunctions on charge separation, carrier lifetime, and surface catalytic reactions has, for the most part, been overlooked. Herein, we develop a tunable defect-level-assisted strategy to construct an electron reservoir, effectively prolonging the lifetime of charge carriers through the rapid capture and gradual release of photoelectrons within WO3-x/In2S3 S-scheme heterojunctions, as authenticated by femtosecond transient absorption spectroscopy and theoretical simulations. The surface photoredox mechanism, unraveled by Gibbs free energy calculations, demonstrates that oxygen-vacancy-induced defect states in WO3-x/In2S3 heterojunctions unlock the rate-determining H2O oxidation into free oxygen molecules by forming metastable oxygen intermediates, contributing to the facilitation of H2O photooxidation. This distinct role, combined with the extended carrier lifetime, results in boosted CO2 photoreduction with nearly 100 % CO selectivity in the absence of any photosensitizer or scavenger. Our work sheds light on the role of controllable defect levels in governing charge transfer dynamics within S-scheme heterojunctions, thereby inspiring the development of more advanced photocatalysts for artificial photosynthesis.

Energy Band Alignment and Defect Synergistic Regulation Enable Air-Solution-Processed Kesterite Solar Cells With the Lowest VOC Deficit.

The major challenge in preparing high-performance Cu2ZnSn(S,Se)4 solar cells is the large open circuit voltage deficit (VOC-def). A new strategy utilizing the synergistic substitution of Ag and In dual cations has been proposed to simultaneously address the problems of undesirable interface band alignment and high-density detrimental bulk defects, obtaining decreased carrier recombination rate and increased minority carrier lifetime. The shorter In-S/Se bonds move the CBM higher by generating stronger repulsive force than the Sn-S/Se bonds, thus adjusting the interface band alignment. Ag substitution can effectively suppress Cu─Zn disorder, while Ag, In dual substitution can further passivate Sn-related defects and solve the issue of low carrier concentration in Ag single-substituted samples. Besides, the superior carrier property of In-Se materials significantly enhanced the device carrier lifetime and minority carrier diffusion length. The state-of-the-art air-solution-processed CZTSSe device without any addition treatment with 14.33% efficiency and 580mV VOC is obtained, possessing the lowest VOC-def value currently available in the CZTSSe field (VOC/VOC SQ = 64.7%). This work emphasizes the synergistic modulation of band alignment, defect level, grain growth, and carrier transportation by dual cation substitution, which paves a convenient and effective way to realize high-performance solar cells and photovoltaic devices.

Constructing surface oxygen vacancy in the [Bi2O2]2+ layer defects mediated Bi2MoO6 enhanced visible light responsive photocatalytic activity.

Bi2MoO6 nanospheres with surface oxygen vacancies (SOVs) controlled by the calcination process were prepared in this study. Performance testing revealed that the Bi2MoO6-4 sample (Bi2MoO6 calcined at 350 °C for 4h) with SOVs achieved a remarkable photocatalytic degradation efficiency up to 99.16% for Rhodamine B (RhB) within 50min, which is 2.19 times higher than that of pure Bi2MoO6. The higher photocatalytic performance of the Bi2MoO6-4 sample is attributed to the SOVs' defect level located at the Bi2MoO6 bandgap, narrowing the bandgap to effectively promote the photogenerated charge separation. The promotion of photocarrier separation and electron were transferred due to the Bi-O bond breakage in the Bi2MoO6-4 [Bi2O2]2+ layer, which mediates the defect level of SOVs in the band structure. The density functional theory calculation results reveal the possible formation site of the oxygen vacancy and the vacancy-induced defect states. This study provides a new approach for fabricating new photocatalysts with surface oxygen defects.

Improving the Classification of Defect Levels in Main Electrical Equipment Through Advanced Prompting on a Large Language Model

The stable operation of the power system is closely related to the national economy and people’s livelihoods. Therefore, the timely detection, qualitative assessment, and handling of major equipment defects are crucial. The classification of defect levels in main electrical equipment is a fundamental task in this process, which is often manually completed, supplemented by knowledge bases or expert systems. However, this approach is time-consuming, labor-intensive, involves challenging human–machine interaction, and relies on expert experience. Conversational large language models, such as ChatGPT, ERNIE Bot, ChatGLM, have garnered widespread recognition in various domains. However, these models may have errors in the reasoning process, resulting in biased or even erroneous outputs, which is referred to as the “hallucinations”. The hallucinations’ problem of large language models poses challenges in specific fields. To mitigate the hallucinations in large language models, researchers often seek to incorporate domain-specific knowledge into these models through methods like fine-tuning or prompt learning. In order to enhance model performance while minimizing computational costs, this study adopts the prompt learning approach. Specifically, we propose a large language model prompt learning framework based on knowledge graphs, aiming to provide the large language model with reasoning support by leveraging specific information stored within the knowledge graph and receive explainable reasoning result. Experimental results demonstrate that our module achieved superior results on the power defect dataset compared to the non-prompt method.

Spraying-Deposited Transparent p-Type Sn-Doped CuI Film and Its Ultrahigh-Speed Self-Powered Photodetector.

The exploitation of simply processed p-type semiconductors and photodetectors with promising optoelectrical properties remains challenging yet essential for current and future advanced optoelectronic applications. Transparent p-type CuI and Sn-doped CuI (Cu-Sn-I) films and their self-powered photodetectors have been successfully fabricated by the spraying method. It is found that the incorporation of Sn dopants enhances the optical, electrical, and photoelectric properties of CuI thin films as well as their corresponding self-powered heterojunction photodetectors. This improvement of the optoelectrical properties of the Cu-Sn-I film and its photodetector can be attributed to the adjustment of the acceptor defect level and increased hole concentration resulting from Sn doping. The Cu-Sn-I/n-Si photodetector exhibits a responsivity of 10.7 mA/W, a detectivity of 6.79 × 1011 Jones, and a response time of 77 μs/30 μs (0 V bias). The response time exhibits the fastest rise and decay times compared with the other CuI-based self-powered UV photodetectors in recent years, showcasing promising applications in the realm of transparent electronics moving forward. This study also presents an effective strategy for enhancing the electrical properties of p-type semiconductors and devices through effective doping.

Classification of defect levels of power equipment based on improved BERT

Abstract Accurate defect level classification in power systems is crucial for preventing outages and ensuring safety. Traditional methods rely on manual inspection, which is inefficient and highly subjective, leading to potential misjudgments. To address this issue, this paper proposes the AMF-BERT model, which dynamically integrates multi-layer output features of the BERT model through an attention mechanism, capturing semantic information at different levels and significantly improving classification accuracy and robustness. The study also analyzes the impact of feature fusion at various layers on defect category classification accuracy. Extensive experiments on power defect text validate the model’s effectiveness, achieving an F1 score of 92.11%, which is an improvement of 3.58% over the baseline BERT model. The results indicate that AMF-BERT outperforms mainstream text classification models in accurately identifying major and urgent defects. This research enables power companies to allocate resources more efficiently, respond swiftly, and reduce safety risks and economic losses caused by improper defect handling, highlighting its practical application value in smart grid management.

Optimisation of Railway Tamping Scheduling

This research was devoted to optimising opportunistic tamping scheduling to present a cost-effective approach that considers both preventive and corrective tamping activities. To achieve this, we formulated the track geometry tamping scheduling problem as a mixed integer linear programming model and employed a genetic algorithm for its resolution. Key track quality indicators, including the standard deviation of the longitudinal level and single defects, were considered.We developed predictive models for the evolution of standard deviation and single defects over time, which were utilised to schedule preventive tamping activities and anticipate potential corrective actions. Additionally, we investigated the impact of both preventive and corrective tamping activities on the values of standard deviation and single defects.A case study on data from the Main Western Line in Sweden demonstrated that the fixed cost of occupying each maintenance window significantly influenced the total tamping cost. Moreover, the maintenance cycle interval notably affected the number of required corrective tamping activities. Specifically, a 3-month interval led to over 50% fewer corrective tamping activities when compared to a 9-month interval. The results revealed that a 6-month interval achieved a favourable balance between corrective and preventive tamping activities and the total cost in our case study.

Synthesis of structure controllable carbon aerogel with low drying shrinkage and Scalable size as High-Temperature thermal insulator

Carbon aerogels, owing to the high-temperature durability and low thermal conductivity, are promising candidates for high-temperature thermal insulation used in inert atmospheres. However, significant drying shrinkage poses challenges in fabricating large-sized carbon aerogels and optimizing their thermal insulation properties. Herein, we report the synthesis of monolithic carbon aerogels with low drying shrinkage and controllable nanostructures by adding self-crosslinked chitosan to provide reaction sites and three-dimensional network support based on the traditional resorcinol–formaldehyde method in an acetic acid environment. Owing to the increased cross-linking degree of resorcinol–formaldehyde gel from the polymerization and hydrogen bonding between chitosan and resorcinol–formaldehyde oligomers, the skeleton of gel network is strengthened and the shrinkage during supercritical drying is as low as 1.51 %, resulting in a good monolithic shape of carbon aerogel (130 × 130 × 15 mm). Besides, they exhibit typical nanoporous characteristics with uniform and small particle size (9–130 nm) and low density (0.070–0.221 g·cm−3), achieving a low thermal conductivity of 0.047 W·m−1·K−1 at 1100 °C and high compressive strength of 1.80 MPa (density of 0.138 g·cm−3). This method facilitates the production of carbon aerogels with low shrinkage, minimal defects, low density, and small pore size, rendering them suitable for applications requiring large dimensions and low defect levels, such as fiber-reinforced composites and high-performance thermal insulators.

Selection of dopants and doping sites in semiconductors: the case of AlN

The choices of proper dopants and doping sites significantly influence the doping efficiency. In this work, using doping in AlN as an example, we discuss how to choose dopants and doping sites in semiconductors to create shallow defect levels. By comparing the defect properties of CN, ON, MgAl, and SiAl in AlN and analyzing the pros and cons of different doping approaches from the aspects of size mismatch between dopant and host elements, electronegativity difference and perturbation to the band edge states after the substitution, we propose that MgAl and SiAl should be the best dopants and doping sites for p-type and n-type doping, respectively. Further first-principles calculations verify our predictions as these defects present lower formation energies and shallower defect levels. The defect charge distributions also show that the band edge states, which mainly consist of N- s and p orbitals, are less perturbed when Al is substituted, therefore, the derived defect states turn out to be delocalized, opposite to the situation when N is substituted. This approach of analyzing the band structure of the host material and choosing dopants and doping sites to minimize the perturbation on the host band structure is general and can provide reliable estimations for finding shallow defect levels in semiconductors.

Study of the effect of CdCl2 introduction on the high temperature activation treatment of CdTe:As polycrystalline thin films

V-doped CdTe polycrystalline films can achieve both doping activation and defect passivation by high-temperature CdCl2 heat treatment, but this requires simultaneous modulation of the amount of CdCl2 introduced to obtain high-quality films. It is found that increasing the CdCl2 introduction does not change the physical phase structure and lattice constant of CdTe:As thin films, but promotes grain recrystallisation, and can promote the formation of A-center, and inhibit the formation of Cd vacancy (VCd) defects, as well as the formation of deep energy level defects. The results provide guidance for the improvement of high-temperature CdCl2 heat treatment of V-doped CdTe polycrystalline thin films.