ABSTRACT The advancement of high‐energy‐density capacitors necessitates dielectric films with superior breakdown strength. Although enhancing crystallinity is a common strategy to improve dielectric breakdown performance, it is not the sole determinant of breakdown strength. This work systematically investigates the effect of annealing (60°C–140°C) on the microstructure and electrical breakdown behaviour of biaxially oriented polypropylene (BOPP) films. Annealing above 110°C is found to not only increase crystallinity and lamellar thickness but also promotes a rearrangement of the amorphous phase. A combination of in situ fourier transform infrared spectroscopy (FTIR), low‐field solid‐state nuclear magnetic resonance (LF‐SSNMR), and Positron annihilation lifetime spectroscopy (PALS) reveals that annealing reduces free volume and significantly suppresses molecular chain mobility in the amorphous phase, as the annealing temperature raise to 140°C, the fractional free volume decreased from 0.15% to 0.09%, and the segment mobility was significantly suppressed. Consequently, the energy storage density has increased from 1.27 J/cm 3 to 2.7 J/cm 3 , and the charge‐discharge efficiency of the 140°C‐annealed film still reached as high as 94.4%, meanwhile, the breakdown filed strength of the metallised 140°C‐annealed film has also increased by 12.6% (≈67 MV/m). These findings demonstrate that amorphous phase mobility including mobility of functional groups, in addition to crystallinity, is a dominant factor governing breakdown behaviour. The constrained chain mobility impedes charge carrier transport and limits energy accumulation, thereby suppressing the initiation of electron avalanche breakdown. This study establishes a new structure–property relationship centred on the mobility of amorphous chains, identifying it as a key descriptor for design of high‐performance dielectric polymers via annealing.
ABSTRACT High‐voltage power semiconductor devices are subjected to repetitive frequency square‐wave pulse voltages during actual operation, making the online assessment of their packaging insulation material degradation challenging. Based on the principle of charge–atom coupling, this paper proposes a novel assessment method utilising molecular vibration acoustic waves induced by forced charge motion at square‐wave pulse edges. By directly employing the rising and falling edges of high‐voltage square‐wave pulses inherent in device operation as excitation sources, this method stimulates molecular chain vibrations within the insulation material and achieves in situ degradation assessment through synchronous detection of acoustic signals. Results indicate that pulse repetition frequency has no significant effect on vibration amplitude; vibration amplitude is proportional to the square of the pulse voltage amplitude and temperature exhibits different effects on the two materials—silicone gel shows a monotonic increase in vibration amplitude with rising temperature, whereas silicone rubber initially increases and subsequently decreases. Under fast edge excitation, low cross‐linking density insulation materials exhibit a double‐peak waveform, whereas high cross‐linking density materials consistently exhibit a single‐peak waveform. This phenomenon reveals intrinsic differences in material chain segment dynamics, providing a new technical pathway for the online monitoring of power device insulation systems without requiring external Gaussian pulse generators.
ABSTRACT Quantifying nanoparticle dispersion remains a central challenge in the development of polymer nanodielectrics, where electrical and dielectric properties are strongly governed by the three‐dimensional (3D) dispersion state of inorganic fillers. This study is aiming to establish a new quantitative framework that enables direct inference of the 3D dispersion state—expressed as the cluster size—from experimentally accessible two‐dimensional (2D) transmission electron microscopy (TEM) images. The framework integrates TEM image analysis, 3D numerical modelling, and machine‐learning regression, thereby linking 2D morphological descriptors to a physically interpretable 3D dispersion parameter. TEM images of three kinds of epoxy nanocomposites with different filler sizes and dispersion conditions were processed to extract particle contours, Feret maximum diameters, and inter‐particle distance distributions, from which a normalised full width at half maximum (FWHM) was obtained as a 2D descriptor of spatial heterogeneity. To clarify the physical origin of this descriptor, a 3D Thomas‐cluster model was constructed to generate particle configurations with controlled dispersion states and filler contents. Cross‐sectional slices extracted from the simulated structures reproduced the experimentally observed relationship between the normalised distance‐distribution width and the underlying 3D dispersion state, represented by the cluster size parameter. Using these simulation‐derived datasets, a LightGBM regression model was trained to predict the 3D dispersion state directly from two TEM‐extracted features: the normalised FWHM and the filler content. The model successfully inferred the relative dispersion states of the experimental samples, demonstrating that these 2D descriptors contain sufficient information to estimate the underlying 3D organization of nanoparticles. The proposed framework provides a physically grounded, data‐driven methodology for quantitative dispersion assessment and establishes cluster size as a robust, transferable metric for describing nanoparticle dispersion in polymer nanodielectrics.
ABSTRACT High‐crystallinity polypropylene (PP) has broad applications due to its excellent mechanical properties, thermal stability and processing properties. However, its capacitive energy storage performance deteriorates severely under high electric fields. In this paper, two surface modification methods, ultraviolet irradiation and surface fluorination, were adopted to modify the commercially biaxially stretched polypropylene (BOPP) film, and the effects of these methods on the surface structure, chemical composition, thermal properties, mechanical properties, dielectric properties and energy storage properties of the material were systematically studied. The results show that the introduction of oxygen‐containing polar groups by ultraviolet irradiation and the formation of C‐F bonds through fluorination treatment can effectively augment the electron injection barrier, suppress charge injection and migration and significantly improve breakdown strength, energy density and charge–discharge efficiency. The breakdown strength was significantly improved from 750 MV/m of pristine BOPP film to 790 MV/m of UV irradiated film and 831 MV/m of fluorinated sample, respectively. As a result, the energy density of the ultraviolet irradiated sample is 5.62 J/cm 3 and the efficiency is 92%. The energy density of the fluorinated sample can reach 5.9 J/cm 3 and the efficiency is 93%.
Bismuth ferrite (BiFeO3, BFO) is a prominent lead-free multiferroic material, whereas strontium titanate (SrTiO3, STO) is a high-permittivity perovskite oxide widely used to tailor dielectric performance. In this work, the solid-solution formation of (1 - x)BFO-xSTO (0.1 <= x <= 0.4) nano-ceramics was synthesised via a low-temperature molten salt method using NaCl as a flux. Rietveld refinement of X-ray diffraction patterns confirmed the formation of a single-phase perovskite structure without secondary phases. The FESEM micrographs indicated the uniform granular-shaped grain morphology of BFO. The XPS revealed the presence of mixed-valence Fe ions, indicating partial reduction within the BFO lattice. The incorporation of STO suppressed oxygen vacancies, enhancing the dielectric constant (similar to 151 at 100 kHz for x = 0.4) and reducing the dielectric loss with increasing STO content. M-H loops exhibited weak ferromagnetic behaviour, attributed to enhanced spin canting and lattice distortion. These results highlight the effective tuning of dielectric and magnetic properties through STO substitution, demonstrating the potential of BFO-STO ceramics for multifunctional electronic device applications such as spintronics and dielectric storage components.
The rapid advancement of new energy vehicles has exposed critical limitations in conventional enamelled wire insulation materials for drive motors, particularly in meeting escalating operational demands. Polyimide is widely adopted in the motor insulation systems, and its inherent corona resistance remains insufficient under extreme conditions. Herein, we propose a strategy to improve the corona resistance of PI films or PI polymer based on the integration of silicon dioxide (SiO 2 ) and aluminium nitride (AlN) nanoparticles. The results indicate that the excellent insulation of SiO 2 and the high thermal conductivity of AlN can lead to a strong effect in improving the corona resistance life of PI. The resultant polymer film (MPI/ASA3) exhibits an excellent corona resistance life of 184.7 min which is 23.38 times higher than that of the MPI/1.0 vol% AOC film. Meanwhile, it still maintains excellent thermal and mechanical properties. Hopefully, our work could promote the advancement of the drive motor for new energy vehicle technology.
Perfluoropentanone (C 5 F 10 O) gas mixtures are expected to be used in gas‐insulated switchgear due to their excellent environmental and insulating properties. Nevertheless, the study of the insulating properties and influencing factors of C 5 F 10 O gas mixtures under the internal conditions of gas‐insulated switchgear is not yet comprehensive. This paper comprehensively investigates the insulating properties of C 5 F 10 O/dry air gas mixtures. The findings demonstrate that adding C 5 F 10 O significantly enhances the insulating properties of the gas mixture. The findings of this research serve as a valuable reference for the engineering application, operation, and maintenance of C 5 F 10 O/dry air gas mixtures.
A series of innovative electron donors (D1-D6) featured both internal and external electron donors in one molecular structure were designed and synthesised. The synthesised electron donors and traditional electron donors (DIBP) were effectively reacted with titanium-based catalysts supported on magnesium chloride to obtain a series of Ziegler-Natta catalysts (Cat 0-Cat 6). Cat 0 formed by electron donor DIBP. Similarly, Cat 1 by D1, Cat 2 by D2, Cat 3 by D3, Cat 4 by D4, Cat 5 by D5 and Cat 6 by D6, respectively, which were characterised by GC-MS, H-1 (C-13)-NMR, XPS and SEM. It is interesting that the catalytic activity of the catalysts prepared by the newly synthesised electron donor was significantly higher than that of traditional DIBP for propylene polymerisation. Under atmospheric pressure condition of propylene, the catalytic activity of Cat 3 was 112.7 g PP/(g Cat h) and Cat 6 reached 209.2 g PP/(g Cat h) at the same conditions, both significantly higher than that of 34.1 g PP/(g Cat h) of Cat 0. We adopted a prealkylation strategy for the catalysts preparation, reducing the optimal aluminium titanium ratio for propylene polymerisation from 50 to 30, decreasing the amount of cocatalyst used and thus reducing the ash content in the products. During the bulk polymerisation of propylene, the activity of prealkylated Cat 3 and Cat 6 is comparable, at 79.4 kg PP/(g Cat h), which is about 2.3 times that of Cat 0 cooperated with external electron donor C during the polymerisation processes (35.2 kg PP/(g Cat h)), and the obtained PP isotacticity reaches over 98%. Especially, Cat 1-Cat 6 were required no external electron donor for propylene polymerisation. In addition, the theoretical ash content of the products formed by Cat 3 and Cat 6 is only 22 ppm, which is significantly lower than that of Cat 0 (133 ppm). It is expected to be used in the industrial production of ultra-clean iPP powder for capacitor films.
As a critical component in transformer partial discharge monitoring and localisation, the output signal of the contact ultrasonic sensor is significantly influenced by its sensitivity. Consequently, field sensitivity verification of ultrasonic sensors is essential. However, field calibration is often impacted by environmental noise, and the effects of various noise types on the sensitivity calibration results remain inadequately understood. To address this issue, this paper introduces a sensitivity field calibration process tailored for ultrasonic sensors used in oil-immersed transformer. Transformer and ultrasonic sensor models are developed, and the sensor output signal is simulated using the finite element method. Noise signals of different types are superimposed to evaluate their effects on the sensitivity calibration results. The results indicate that under the interference of narrowband noise, white noise, and mixed noise, significant errors occur in both the root mean square fluctuation of the sensitivity curve and the peak sensitivity, whereas the error in mean sensitivity is comparatively smaller than that of the aforementioned indicators. These results provide a theoretical foundation for the development of targeted denoising techniques during field verification.
The composite cermet material based on BaTiO3 ceramics doped with Fe was compacted using the field-assisted sintering technology (FAST). Certain oxidation of Fe and also the presence of carbon, used in sintering equipment, were detected in the sintered material. The microstructure showed well separated two components and certain porosity. Dielectric parameters were measured between 30 degrees C and 150 degrees C and between frequencies 100 Hz and 20 MHz. Combination of dielectric parameters at 1 MHz frequency, that is, relative permittivity 5 200, loss tangent 0.051 and thermal coefficient of capacitance 715 ppm/degrees C made this composite promising among the addressed category of materials. The mechanisms of polarisation and conduction and their thermal activation were discussed.
This study investigates the physical-chemical and electrochemical performances of innovative separators for Li-ion batteries based on poly(vinylidene fluoride-hexafluoropropylene) (PVDF-HFP) nanofibrous membranes. The nanofibrous mat is produced through the electrospinning process, ensuring high surface to volume (S/V) ratio and allows large-scale production under suitable conditions. The materials investigated in this work aim at overcoming some limitations specific to the commercial separators, for example, mechanical shrinkage and electrolyte uptake. This is achieved by adding nanoparticles of different types, for example, ZrO2, SnO2, SiO2 at different concentrations. Results claim that nanofibrous separators improve the mechanical and thermal stability of the mat without significantly impacting its electrochemical performances. In particular, the addition of 7 wt% of ZrO2 nanoparticles to the nanofibrous separator showed to outperform commercially available solutions (i.e., Celgard) in terms of mechanical and thermal stability exhibiting, also, electrochemical performances.
Metallised film capacitors consist of polymer dielectrics and electrodes which are considered as two layers. Due to the huge difference in thickness of two layers, metallisation can affect the microstructure and properties of metallised films, resulting in the bilayer structure model may not accurately describing properties. In this work, metallised films are considered as a whole rather than two layers of dielectric and electrode, whereas the process of film metallisation is regarded as a surface modification of the dielectric film. Metallised films with different metal layer thicknesses are prepared by vacuum evaporation. We investigated the dielectric, electrical and self-healing properties of metallised films. Films with thicker electrode present higher dielectric constant, conductivity, energy density and lower breakdown strength. Thinner metal layers result in less self-healing energy and better self-healing property. The difference in properties is attributed to the surface modification of dielectric films by nanoscale metallisation. Thicker electrode contributes to smaller surface roughness, which increases the polarization and charge capacity of metallised films, but also implies worse self-healing property. Surface modification provides a new perspective for researching metallised films.
The development of a reliable, environmentally safe, and economic insulating oil for the transformer is an endless effort of the electrical industry. Mineral oil (MO), traditionally used, presents challenges including high cost, environmental impact, and limited availability. Consequently, research has shifted towards alternative solutions, aligning with green energy and environmental conservation goals. Natural ester oil has gained keen attention due to its complete biodegradability and widespread availability. This study explores alternative options, including a blend of vegetable oils (Blackseed, Castor, Flaxseed, and Mustard), aiming to enhance transformer insulation and cooling efficiency through this novel nanofluid. A novel nanofluid was synthesised using different ratios of proposed oils to produce a 300 mL sample blend of vegetable oil containing 0.03 g (0.0001 wt%) of green-synthesised SiO2 nanoparticles (average size similar to 20-30 nm). These nanoparticles were fabricated using an eco-friendly method based on Moringa leaf extract, aligning with Sustainable Development Goals (SDGs). This sample was prepared for testing and validation at the advanced High Voltage Laboratory in accordance with the IEC-60156 standard to evaluate dielectric properties. Furthermore, thermal stress was applied to the proposed samples to replicate real-time conditions. Moreover, conducted comparative analyses between a newly proposed vegetable oil-based nanofluid and conventional MO, emphasising dielectric properties, including AC and DC breakdown voltage (BDV) and stability. Results demonstrate that the eco-friendly vegetable oil-based nanofluid surpasses traditional MO by a significant 19% margin in breakdown strength. Stability assessments reveal only a 0.32% reduction in BDV for the proposed nanofluid after 6 years of simulated equivalent ageing, contrasting with a 19% reduction observed in MO. This contemporary research highlights the potential of the proposed nanofluid as a promising alternative to conventional MO, as evidenced by comprehensive testing and analysis.
The emergence of polymer nanodielectrics as suitable materials in energy storage devices highlights the importance of a deep understanding of their dielectric/electrical properties. When conductive materials are employed as nanofillers, strong interfacial polarisation and free charge carrier transport phenomena occur that both contribute to the imaginary permittivity as a peak and as a power law in the form of sigma 0/(epsilon 0 omega s ), respectively. To effectively discern and understand the two phenomena, the use of different dielectric formalisms is often preferred, that is, by the complex electric modulus M*(omega). However, when M*(omega) is employed, the free charge carrier contribution transforms into a step in M '(omega) and into a peak in M ''(omega), namely, the conductivity relaxation, that can be difficult to distinguish from interfacial polarisation or dipolar effects. A general relation is proposed here to describe the non-Debye polarisation component of interfacial polarisation and non-Ohmic transport of free charge carriers in its electric modulus representation. From this relation, the conductivity relaxation can be isolated into a separate function and is shown to exhibit a symmetrical broadening that cannot be described by the semi-empirical Havriliak-Negami function when it deviates from Ohmic behaviour.
In the high-humidity environment, the epoxy resin (EP) insulating materials commonly used in reactors and other equipment are prone to accelerate ageing due to the damp and hot conditions, which affects the long-term stable operation of the equipment. However, at present, the research on the blocking effect of carbon nano-modified fillers on the invasion of water molecules into the epoxy resin matrix is still unclear. Based on this, we utilised molecular dynamics simulation technology to establish six EP models of EP materials under different conditions, including water molecule invasion (WI)/non-invasion (WNI) and carbon nanomaterial modification/no modification. The carbon nano-modified materials adopted Amido-Amine Functionalised Carbon Nanotube (AFCNT) and Hydroxyl Functionalised Graphene (HFGNP). We studied the changes in key performance indicators such as relative permittivity, thermal conductivity, and glass transition temperature of six models. Eventually, it was found that the relative permittivity of the EP model significantly increased after water molecule invasion but decreased after doping with AFCNT. The thermal conductivity of the six EP models increased, and the increase in the AFCNT-EP model was the most obvious, rising by 46.65% at 300 K. The glass transition temperatures of the models all decreased, but the decrease was reduced after doping and modification with carbon nanomaterials. Overall, except for thermal conductivity, the overall performance of the EP model deteriorated to varying degrees after water molecule invasion. However, the performance deterioration trend of the model was alleviated after doping with carbon nano-modified fillers, and the performance of the model doped with AFCNT was significantly better than that of the EP model doped with HFGNP. This research can provide certain theoretical basis and technical support for the engineering application of epoxy resin materials under extreme conditions such as high humidity.
In this study, a magnetic disk was prepared using nanoparticles with a diameter of less than 15 nm. The morphological and structural characteristics of these nanoparticles were systematically examined using X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and alternating force gradient magnetometry (AGFM). XRD analysis confirmed that the average diameter of the copper-magnesium ferrite nanoparticles doped with cadmium was approximately 12 nm, consistent with TEM results, which also showed uniform particle distribution and a tendency to form clusters in powdered form. AGFM measurements revealed that the magnetic property of the powder sample was 15.83 emu/g, which increased to 22.70 emu/g after compression, highlighting the influence of particle density and morphology on magnetic behaviour. Gas sensing tests demonstrated that the fabricated sensors achieved exceptional sensitivity, particularly to acetonitrile, with a maximum sensitivity of 92.3%. A hybrid deep learning model, Bi-LSTM, was utilised to enhance the precision of gas classification. The proposed methodology was benchmarked against traditional machine learning models, including LSTM and RNN, and demonstrated superior performance. The accuracy of gas detection reached an impressive 99.89%, as validated by ROC analysis, underscoring the efficacy of the deep learning-based approach. These findings highlight the potential of cadmium-doped ferrite nanoparticles for high-performance gas sensing applications, suitable for both industrial and medical uses.
Natural ester insulating oils are increasingly replacing mineral oils as the insulating medium for oil-filled equipment due to their high ignition point, biodegradability, and other environmentally friendly properties. However, the reaction characteristics of natural ester insulating oils under different types of faults require further investigation. This paper presents the development of a molecular dynamics model employing the ReaxFF reactive force field to comprehensively simulate the decomposition of natural ester insulating oils over a temperature range of 2800-4000 K, elucidating the resulting product information. The gas production behaviour of natural ester insulating oils was examined under different overheating conditions and heating times. The simulation results indicate that the thermal decomposition products of natural ester insulating oil primarily consist of seven gases, including H2, CO, CO2, and others. Notably, the concentration of C2H4 gas exhibits a significantly nonlinear negative correlation with overheating conditions. Because of variations in simulation temperatures, the characteristic gases generated during the thermal decomposition of natural ester insulating oil correspond to different fault types observed in real-world scenarios. Specifically, the gas production at a simulated temperature of 4000 K aligns with the gas production behaviour of insulating oil during discharge fault events in practice. The results of this study offer a theoretical basis for the application of insulation condition monitoring in oil-filled equipment through dissolved gas analysis (DGA).
Silicone rubber (SIR) composite insulators are widely employed in electrical applications due to their exceptional chemical stability, low surface energy and superior electrical insulation properties. To enhance the hydrophobicity and low-temperature resistance of SIR, blended composites with varying ratios of SIR and phenyl silicone rubber (PSIR) were fabricated. The study revealed that matrix-filler network interactions between PSIR and fillers were weaker compared to those in SIR-based systems. Increasing PSIR content led to reduced elongation at break in the composites, while tensile strength remained largely unchanged. Concurrently, the breakdown strength is inferior to that of pure PSIR composites. Notably, the blend of SIR and PSIR enhances both hydrophobicity and resistance to hydrophobicity migration. This work provides a strategic approach for enhancing the performance of SIR composites suitable for applications in regions with high humidity and significant rainfall.
Epoxy resin-based dielectric materials are widely used for electrical insulation. However, due to the application of electrical equipment in complex and diverse climatic environments, the characteristic variation of the material must be paid attention to. Besides, decoupling of the overall performance is necessary to trace the source of property variation. Thus, in this research, the polarisation characteristics over broad ranges of temperature and frequency of typical epoxy resin used for electrical insulation are investigated. Moreover, the Dissado-Hill model is adopted to quantitatively decouple the individual contribution of each microprocess. It indicates that the magnitudes of the polarisation processes below the glass transition temperature (T g) are small, the influence of which on the overall polarisation characteristics is marginal. In comparison, the influence of alpha relaxation initiating above T g on the magnitude of permittivity and the dielectric loss is prominent, which must be fully taken into consideration during insulation design. In addition, due to the long structure of repeating units in the backbone of epoxy, another polarisation process would appear at higher temperatures stemming from the internal relaxation of repeating units. Furthermore, the amplitude of conductance increases exponentially as temperature rises and would become the major component of loss at elevated temperatures. This research provides insights into the rational design of insulation structures and the development of novel epoxy materials.
During the operation of HVDC transmission systems, the oil-pressboard insulation of converter transformers is constantly subjected to combined AC and DC voltages. The DC voltage component can easily lead to charge accumulation and induce surface flashover. To investigate the development process of surface flashover in oil-pressboard insulation under DC voltage and its influencing factors, this paper establishes a two-dimensional simulation model using finite element simulation software to study the electric field and charge distribution characteristics of oil-pressboard composite insulation systems. Using the established model, the paper analyses the influence of voltage, gap distance and electron mobility on the surface flashover development process.
