Papers
General and interdisciplinary, including biophysics, geophysics and astrophysics
035001
The following article is Open access
Yoshiyuki Hirano, Kanta Nambu, Tsukasa Aso, Masanori Hara and Susumu Fujiwara
Ionizing radiation induces DNA damage both directly and indirectly through hydroxyl radicals (·OH) generated by water radiolysis. The indirect action involves hydrogen abstraction from deoxyribose, leading to DNA damage such as base release and strand breaks. The probability of such events depends on the accessibility of specific hydrogen atoms to ·OH radicals. Previous computational studies have examined naked DNA fragments; however, the effect of nucleosomal organization has not been addressed. Here, we performed molecular dynamics (MD) simulations using LAMMPS (Large-scale Atomic Molecular Massively Parallel Simulator) to evaluate the accessibility of ·OH radicals to deoxyribose hydrogens in both 12-bp naked B-type DNA and nucleosomal DNA containing histone proteins. The accessibility to the hydrogens increased, following the order H1′ ≈ H2′ ≈ H3′ < H4′ < H5′, consistent with previous studies. Accessibility was approximately proportional to the solvent accessible surface area (SASA), supporting SASA as a practical predictor of accessibility even in base-dependent differences were observed. In nucleosomal DNA, accessibility of hydrogens facing the histone core was significantly reduced compared to solvent-facing hydrogens, despite comparable SASA values. This reduction suggests that histones physically hinder ·OH radical penetration. These findings highlight the importance of considering protein–DNA complexes when modeling indirect radiation effects.
035002
The following article is Open access
Yahia Al-Omar, Majida Nahili and Nidal Chamoun
We present a unified framework that combines early- and late-Universe observations to constrain three functional realizations of
gravity: the linear, quadratic, and general power-law models. First, constraints on deviations from the standard weak interaction freeze-out temperature are derived using the most recent measurements of the primordial helium-4 mass fraction. Second, we perform a joint analysis incorporating five priors: Type Ia supernovae, baryon acoustic oscillations, cosmic chronometers, Big Bang Nucleosynthesis, and Cosmic Microwave Background in order to place bounds on the model parameters. The joint likelihood analysis significantly tightens the constraints compared to individual datasets. Third, we test the null, strong, and dominant energy conditions to evaluate the physical viability of the best-fit solutions across the cosmic redshift range. Our results show that all three
models are consistent with current observations and exhibit stable behavior under the energy-condition criteria, supporting torsion–boundary modified gravity as a robust and viable alternative of General Relativity.
035003
Haochen Zheng, Xueqiang Ji, Jinjin Wang, Jianying Yue, Gang Liu, Weihua Tang and Peigang Li
High surface fields represent an inherent trade-off for Ga2O3 power devices concerning their on-state energy consumption and voltage capability. This work proposes a Ga2O3 Schottky barrier diode that incorporates an embedded trench Schottky junction and a metal-oxide-semiconductor (MOS) capacitor. This structure aims to reduce surface field and mitigate the peak electric field, thus achieving low power consumption and high breakdown voltage. Leveraging the trench design, the T-SBD achieves a breakdown voltage of 571 V, double that of a conventional SBD (285 V). The trench-wrapped electrodes enhance sidewall depletion under reverse bias, while the MOS capacitor grid columns further suppress leakage current to as low as 8.6 μA cm−2. Furthermore, TCAD simulation results further demonstrate the effective electric field optimization of the devices with Trench SBD structure. This research lays a solid foundation for the development of higher-power Ga2O3 Schottky diodes.
035004
Ashish Kumar, Ashok Kumar Bagha and Sumit Sharma
Microwave hybrid heating (MHH), which combines microwave irradiation with conductive heating from a susceptor, provides an efficient and energy-saving approach for metal melting. In this study, microwave-assisted melting and casting of metallic powders (aluminium, copper and nickel) was carried out in a 900 W, 2.45 GHz household microwave oven using a graphite crucible and activated charcoal as hybrid susceptors. A comprehensive multi-physics simulation framework was developed to model and optimize the MHH process of metallic powder within the standard microwave oven. This framework integrates electromagnetic field distribution, heating losses, and thermal profiling to capture the complex interactions between microwaves, susceptors, and metallic powder during heating. A simulation study was conducted to predict the exposure time required for efficient melting, providing valuable insights for process optimization. The simulation model showed a close correlation with the experimental results, with an average percentage error of 7.71% in predicting exposure time for different metallic materials across different crucible geometries. Additionally, the simulated thermal profile of the susceptor domain was compared with experimental observations to evaluate model accuracy and reliability. By enabling process optimization prior to physical experiments, the simulation saves time and resources, reduces energy consumption and material waste, and supports the goals of Industry 4.0 for sustainable and efficient manufacturing.
Quantum optics, processing and information
035101
The following article is Open access
Nacer Eddine Belaloui, Abdellah Tounsi, Abdelmouheymen Rabah Khamadja, Hamza Benkadour, Mohamed Messaoud Louamri, Achour Benslama and Mohamed Taha Rouabah
The inherent noise in current Noisy Intermediate-Scale Quantum (NISQ) devices presents a major obstacle to the accurate implementation of quantum algorithms such as the Variational Quantum Eigensolver (VQE) for quantum chemistry applications. This study examines the impact of error mitigation strategies on VQE performance. We show that, for small molecular systems, an older-generation 5-qubit quantum processing unit (IBMQ Belem), when combined with optimized Twirled Readout Error Extinction (T-REx), achieves ground-state energy estimations an order of magnitude more accurate than those obtained from a more advanced 156-qubit device (IBM Fez) without error mitigation. Our findings demonstrate that T-REx, a computationally inexpensive error mitigation technique, substantially improves VQE accuracy not only in energy estimation, but more importantly in optimizing the variational parameters that characterize the molecular ground state. Consequently, state-vector simulated energies suggest that the accuracy of the optimized variational parameters provides a more reliable benchmark of VQE performance than quantum hardware energy estimates alone. Our results point to the critical role of error mitigation in extending the utility of noisy quantum hardware for molecular simulations.
035102
Anas M Salhab
In this paper, we consider an efficient dual-hop quantum communication network with multiple senders or nodes, a repeater, and a receiver. We derive closed-form expressions for the system quantum outage probability (QOP) assuming time-varying amplitude damping (TVAD) quantum channels. In addition, closed-form expressions for the quantum hashing outage probability of two approximated quantum channels, namely TVAD Pauli twirl approximated (TVADPTA) and TVAD Clifford twirl approximated (TVADCTA) channels are also provided. These channels are efficient in modeling the simplest kind of noise in quantum devices. Opportunistic node scheduling is proposed, for the first time, to select among sending nodes, where the node of the largest relaxation time T1 is allowed to transmit to repeater, which in turn communicates with receiver. The impact of various system parameters, including number of sending nodes K, coefficient of variation of relaxation time
, code rates of both hops, channel damping parameter γ, and channel noise limit γl(R) on the system performance is investigated. Findings show that for the case where first hop is dominating the system performance, the more the number of nodes, the better the achieved performance. This gain starts to vanish once the second hop starts dominating the system performance. For good quality channels, more nodes are needed in order for second hop to dominate the system performance compared to the case of low quality channels. Finally, results illustrate that for higher values of
, and in order to achieve a certain target QOP, both nodes and repeater need to adapt their transmission rates to the optimum values.
035103
Seokwon Choi, Talal Ahmed Chowdhury and Kwangmin Yu
We implement a quantum error mitigation method termed self-mitigation, which is comparable to zero-noise extrapolation, at large scales to achieve quantum utility on near-term, noisy quantum computers. We investigate the effectiveness of several quantum error mitigation strategies, including self-mitigation, by simulating quantum quench dynamics for Heisenberg spin chains with system sizes up to 104 qubits using IBM quantum processors. In particular, we discuss the limitations of zero-noise extrapolation and the advantages offered by self-mitigation at large scales. The self-mitigation method demonstrates stable accuracy with large systems of 104 qubits comprising more than 3,000 CNOT gates. Also, we combine the discussed quantum error mitigation methods with practical entanglement entropy measuring methods, and it shows a good agreement with the theoretical estimation. Our study illustrates the usefulness of near-term noisy quantum hardware in examining the quantum quench dynamics of many-body systems at large scales and lays the groundwork for surpassing classical simulations with quantum methods prior to the development of fault-tolerant quantum computers.
035104
Zhen-Zhen Li, Ming-Kui Liu, Wen-Ling Yang, Bo Gao and Zi-Chen Li
Quantum homomorphic encryption is the corresponding technology of classical homomorphic encryption in the quantum field. Due to its ability to ensure the correctness of computation and the security of data, it is particularly suitable for delegated computation in quantum cloud networks. However, previous schemes were unable to simultaneously handle the volatility problem of servers dynamically and eliminate the error caused by homomorphic evaluation of T-gate non-interactively. Therefore, a multi-party dynamic quantum homomorphic encryption scheme based on rotation operators is proposed in this paper. Firstly, the proposed scheme uses the rotation operators to solve the phase gate error that occurs during the homomorphic evaluation of T-gate non-interactively. Secondly, the scheme can dynamically deal with instability of servers, such as adding a server or removing a server. Then, the trusted key center is introduced, which is responsible for key updating and circuit replacement, which lowers the requirements for quantum capabilities on the client. Finally, this scheme extends the single-client multi-server model to the multi-client multi-server model, making it more suitable for quantum distributed networks and enhancing its practicality. In addition, we theoretically prove the correctness and fully homomorphic property of the proposed scheme, and verify it through the simulation conducted on the IBM Quantum Experience. Security analysis and efficiency analysis further demonstrate that the proposed scheme is information-theoretically secure and possesses high qubit efficiency.
Mathematical physics, complex systems and nonlinear dynamics
035201
Fei Li, Hao Jing and Yonggang Su
Chaos systems, due to their excellent pseudo randomness, unpredictability, ergodicity, and extreme sensitivity to initial conditions and control parameters, are commonly used in physical layer encryption. However, in practical applications, there are issues such as key leakage or insecurity due to the low dimensionality of chaos systems and the lack of effective association between the key and user identity. To address these issues, this paper proposes a physical layer encryption scheme based on human biometric keys and a five-dimensional hyperchaotic system, achieving multi-level encryption at the physical layer. Firstly, a five-dimensional hyperchaotic system with higher complexity is designed, utilizing biometric parameters as the initial conditions to drive the hyperchaotic system to generate chaotic sequences. Secondly, combining channel block interleaving technology, the chaotic sequences are used to randomly permute the positions of the original data, achieving bit-level randomization and recombination. Thirdly, the constellation diagram of the modulated signal is processed in layers, and the chaotic sequences are used to implement dynamic phase rotation of constellation points. Finally, during OFDM modulation, the chaotic sequences are employed to perturb the carrier index. Simulation results show that the proposed scheme not only provides strong security but also has minimal impact on the performance of the communication system. Compared with traditional chaos-based encryption, the high-dimensional chaos system increases the key space, and the biometric binding mechanism enhances the system’s security.
035202
Shaohui Yan, Zihao Guo and Weitao Hu
This study investigates the impact of complexity on the weak signal detection performance of a chaotic synchronization control system. Experimental results demonstrate that as system complexity increases, the weak signal detection performance improves significantly, confirming a positive correlation between complexity and signal detection capability. To further validate this hypothesis, three typical chaotic systems—Lorenz, SH, and Liu—were selected for experimental analysis. The results consistently show that increasing system complexity enhances weak signal detection performance. Based on these findings, a complete hardware circuit implementation of the most optimal chaotic synchronization control system was carried out, and practical weak signal detection experiments were conducted with multiple types of multi-frequency signals. The results indicate that the hardware system not only outperforms other designs in signal detection performance but also features a simple circuit structure, low cost, and strong potential for engineering applications.
035203
Shuaishuai Shi, Licai Liu, Chuanhong Du and Zefeng Zhang
Incorporating memristors into nonlinear systems can significantly enhance their dynamical complexity. However, in certain classes of memristor-based nonlinear systems, linear dependencies among variables persist, which substantially diminish their nonlinear characteristics. To overcome this limitation, we propose a novel memristor model and integrate it into two nonlinear systems that originally exhibit linear relationships among variables, effectively increasing system complexity. MATLAB simulations confirm the memristive behavior of the proposed model, while Multisim circuit simulations validate its physical realizability. By replacing the original memristor models in the two systems with the new model, nonlinear analyses of the four resulting systems demonstrate the elimination of the linear dependencies present in the original systems. Bifurcation and Lyapunov exponent analyses reveal that the new memristor systems exhibit chaotic behavior over a broader parameter range with larger Lyapunov exponents. Furthermore, spectral entropy analysis indicates that the new models achieve higher complexity under certain parameters, making them favorable for secure communication applications. Notably, both memristor-based chaotic systems allow single-parameter control of signal amplitude and polarity, which simplifies engineering implementation using only a sliding rheostat and a single-pole double-throw switch. Finally, Multisim simulations and digital signal processor-based hardware experiments confirm the physical feasibility of the proposed memristor chaotic systems.
035204
Xiang Hu, Yue Kai, Kai Zhang, Yan Liu and Bailin Zheng
The aim of this paper is to investigate the dynamics properties of soliton propagating in optical system governed by the complex Ginzburg-Landau equation with multiplicative white noise. The original equation is first reduced via the traveling wave transformation, leading to the derivation of Gaussian soliton solutions. Subsequently, the Hamiltonian function is constructed, and the equilibrium points of this system are determined. Abundant exact traveling wave solutions are obtained using the complete discrimination system for polynomial method. Both Gaussian soliton solutions and singular soliton solutions are graphically represented. We also investigate the chaotic behaviors of this model by introducing external perturbation terms The corresponding largest Lyapunov exponents are computed to confirm the presence of chaos. This study provides new insights into soliton dynamics in optical systems influenced by white noise.
035205
Puneet Kumar Pal and Dhirendra Kumar
Digital audio communications involve sensitive information that requires secure transmission over unsecured networks. However, traditional encryption methods are costly for real-time applications due to the large size and high correlations in the audio data. Chaos-based encryption offers computational efficiency and resistance to cryptographic attacks. This research proposes a novel SHIELD map, combining sine, Logistic, and exponential functions with modular operations to enhance chaotic behaviour. A two-step confusion process using a novel bit-level operation and the Fisher-Yates algorithm, along with a dynamic diffusion mechanism based on the improved sine-tangent map, strengthens the proposed audio encryption algorithm. The encryption scheme is evaluated against various cryptanalysis methods, demonstrating a 2512 bit keyspace, 7.999 information entropy, 0.003 correlation coefficient, and NSCR and UACI values of approximately 99.60% and 33.50%, respectively. Additionally, the algorithm achieves high fidelity in decrypted audio with a signal-to-noise ratio exceeding 35 dB and perceptual evaluation of speech quality scores ranging from 4.0 to 4.5. The results confirm the proposed method’s robustness, efficiency, and suitability for secure real-time audio transmission.
035206
Erfang Ma
We analyze the nonlinear forward map of electrical impedance tomography (EIT) with the implicit function theorem. We show that the forward map is infinitely differentiable and use its Taylor expansion to quantify the error of this local linearization. We prove that the error of the linearization of second order could be halved with the popular log-transformation of conductivity in some cases. Simulations on a 3D EIT model also justify this theoretical finding.
035207
Tao Xu, Ying Wang and Yaonan Shan
As a representative of the celebrated Heisenberg ferromagnet-type system, the Kuralay equation is investigated by the Darboux transformation (DT) method. To establish the integrability of the system, we first derive its Lax pair and the corresponding N-fold DT in a compact determinant form. Beginning with the mixed backgrounds, the mixed multi-solitons are derived via the resulting DT, and these solutions comprise N-bright-soliton in q component and N-dark-soliton in v component. The general soliton molecules (SMs) are then constructed by applying the velocity resonance conditions to the aforementioned soliton solutions. Furthermore, the general SMs can be classified into three primary categories: pure SMs, SMs interacting with fundamental solitons, and hybrid solutions comprising different types of SMs. Beginning from the non-zero backgrounds, various types of N-breather solutions are also generated, such as Akhmediev breather, Kuznetsov-Ma breather and the general case. Finally, the dynamics of these three kinds of nonlinear localized waves are discussed in detail.
035208
Li-Jin Luo, Feng-Hua Luo, Hua Zhang and Jian-Hua Yang
The calculation of the equivalent complex impedance between arbitrary-stage nodes in cascaded cobweb complex impedance networks represents a challenging issue in the field of circuit network analysis. To address this challenge, this study proposes a systematic approach composed of the following steps: (1) Replace the cascaded cobweb complex impedance network with a cascaded cobweb pure resistive network that maintains the same connection topology; (2) Introduce the concept of ‘negative resistance’ to decompose the cascaded cobweb resistive network into three equivalent sub-networks—referred to as the left, middle, and right sub-networks; (3) Employ an improved recursion-transformation method to calculate the equivalent resistance between the final-stage nodes of the left and right sub-networks; (4) Based on the parallel relationship among the three sub-networks, determine the equivalent resistance between arbitrary-stage nodes of the cascaded cobweb pure resistive network; (5) Substitute the resistance parameters in the derived resistance expression with their corresponding complex impedance parameters to obtain the expression for the equivalent complex impedance between arbitrary-stage nodes of the original cascaded cobweb complex impedance network. Using this method, the equivalent complex impedance between the arbitrary-stage nodes in a 3 × 6 × m cascaded cobweb LC network was successfully calculated. This approach offers an effective and generalizable solution for determining the equivalent complex impedance or resistance between arbitrary-stage nodes in cascaded circuit networks.
035209
The following article is Open access
M Aa Solberg
We apply Lie symmetry analysis of partial differential equations (PDEs) to the Euler–Lagrange equations of the two-Higgs-doublet model (2HDM), to determine its scalar Lie point symmetries. A Lie point symmetry is a structure-preserving transformation of the spacetime variables and the fields of the model, which is also continuous and connected to the identity. Symmetries of PDEs may, in general, be divided into strict variational, divergence and non-variational symmetries, where the first two are collectively referred to as variational symmetries. Variational symmetries are usually preserved under quantization, and variational Lie symmetries yield conservation laws. We demonstrate that there are no scalar Lie point divergence symmetries or non-variational Lie point symmetries in the 2HDM, and re-derive its well-known strict variational Lie point symmetries, thus confirming the consistency of our implementation of Lie’s method. Moreover, we prove three general results that may simplify Lie symmetry calculations for a wide class of particle physics models. Lie symmetry analysis of PDEs is broadly applicable for determining Lie symmetries. As demonstrated in this work, the method can be applied to models with many variables, parameters, and reparametrization freedom, while any missing discrete symmetries can be identified through the automorphism groups of the resulting Lie symmetry algebras.
Particles, fields and nuclear physics
035301
Dzung Tri Tran, Quang Hoang-Minh Pham, Khoa Ngo-Thanh Ho and Khiem Hong Phan
In this work, we study the one–loop–induced decay channel H± → W±γ in the general
gauge within Two Higgs Doublet Models. We analytically verify the gauge invariance (ξ-independence), ultraviolet finiteness, and renormalization-scale independence of the one-loop form factors, thereby confirming the consistency of our calculations. On the phenomenological side, we perform a parameter scan of the Type-I THDM and, based on the allowed parameter space, evaluate the branching ratios of this decay process. Furthermore, we investigate charged Higgs pair production at muon–TeV colliders through the representative processes μ+μ− → H+H− → W+W−hγ and μ+μ− → γγ → H+H− → W+W−hγ, as typical applications of our results. The events for the processes are computed within the allowed parameter regions of the Type-I THDM. The corresponding signal significances are evaluated, including the relevant Standard Model backgrounds. With the high integrated luminosity expected at muon–TeV colliders, our analysis indicates that the signal can be detected with a statistical significance exceeding 2σ for several benchmark scenarios in the viable parameter space of the Type-I THDM.
Atomic and molecular physics
035401
V Jonauskas
Electron-impact ionization process is studied for subconfigurations of the ground (5d46s) and metastable (5d36s2, 5d5) configurations of the W+ ion using the scaled distorted wave (sDW) approximation. The excitation-autoionization (EA) channels corresponding to excitations from the 4f, 5d, and 6s subshells are investigated. Convergence of the EA channels is determined by including the excitations up to shells with the principal quantum numbers n ≤ 30 and orbital quantum numbers l ≤ 5. Study demonstrates a significant contribution from the indirect process (∼20–40%) to the total ionization cross sections for the analyzed configurations. There is a good agreement between experimental data and theoretical results obtained for the 5d46s and 5d36s2 configurations. The obtained results using subconfiguration averaged approach indicate that further level-to-level studies using sDW or more computationally intensive non-perturbative methods should focus on the 5d46s and 5d36s2 configurations aiming to explain available measurements for the W+ ion. Given its relatively low computational cost and reasonable accuracy, the sDW method proves to be a valuable tool for modeling ionization in fusion plasmas.
Electromagnetism, optics and photonics
035501
Huu Nguyen Bui, Dac Tuyen Le, Bui Xuan Khuyen, Thanh Son Pham and Dinh Lam Vu
Control and localization of magnetic fields are critical issues in wireless power transfer (WPT) technology. Managing and localizing magnetic fields minimizes energy losses in regions without a load, thereby enhancing the WPT efficiency through magnetic induction or resonance. In this study, we propose the application of a digital metamaterial (DMM) capable of tuning localized field on the DMM as called the space-modulated reconfigurable region to improve the WPT efficiency. Based on the exceptional flexibility of the DMM slab, magnetic fields can be localized to various receiver coil locations. The experimental results indicate that the WPT efficiency remains relatively stable, ranging from 67% to 50%, for an Rx:Tx coil area ratio of 1:32 as the transfer distance increases from 15 cm to 20 cm. Furthermore, the proposed DMM enables the WPT system to maintain an efficiency above 36% even under a coil misalignment of up to 15 cm at 15 MHz over a transfer distance of 15 cm.
035502
Abhijeet Kumar, S K Sriwas, Mahendra Kumar, Saurabh Singh and Archana Bharti
A compact, low-profile, and flexible circularly polarized antenna is proposed for off-body 5G wearable applications, featuring a unique folded key-shaped parasitic strip integrated on a Cordura textile substrate. The design excites two orthogonal modes using an asymmetrical feed line, folded stub, and slotted partial ground to achieve circular polarization with a 3 dB axial ratio bandwidth of 21.42% (3.8–4.7 GHz), effectively covering 5G NR bands n77 (3.3–4.2 GHz), n78 (3.3–3.8 GHz), and partial n79 (4.4–5 GHz). The antenna exhibits a measured impedance bandwidth of 66.98% from 2.8 to 5 GHz and a peak gain of 5.2 dBic, while maintaining a compact footprint of 0.28λ0 × 0.14λ0 at 2.8 GHz. Fabricated on a flexible textile with excellent mechanical and electromagnetic properties, the prototype demonstrates strong agreement between simulation and measurement, ensuring stable performance under conformal conditions. This design enables high-speed data transfer, robust off-body connectivity, and accurate real-time monitoring, making it highly suited for next-generation wearable 5G systems with strict constraints on size, gain, and SAR compliance.
035503
Jiayu Shen, Hao Tan, Weijie Xu, Yingfei Zhao, Mingzhi Sun, Fei Wang, Tianrun Liu, Guangzhao Zhu, Shixian Sun, Yuanxiang Xu et al
With the growing research on topological insulators (TIs) in recent years, a variety of novel TIs with excellent nonlinear optical properties have been synthesized—such as SnBi2Te4. Its strong absorption in the near-infrared region makes it a promising candidate as a saturable absorber (SA) for utilize in passively mode-locked fiber lasers. In this experiment, SnBi2Te4 was thoroughly characterized, and a nanosheet solution was prepared via liquid-phase exfoliation (LPE). The nanosheets were then deposited onto the tapered region of a tapered optical fiber via the optical deposition method (ODMs) to fabricate the saturable absorber, exhibiting a modulation depth of 6.058% and a saturation intensity of 15.32 MW cm−2. The fiber laser incorporating the SnBi2Te4 SA was able to generate conventional soliton pulses with a central wavelength of 1532.13 nm, with a signal-to-noise ratio of 60.18 dB. By increasing the pump power, 59th-order harmonic mode-locking was successfully achieved. Furthermore, by adjusting the polarization controller (PC), multi-pulse outputs ranging from the 2nd to the 9th order could be flexibly generated, with smooth switching between different pulse states. These results demonstrate that SnBi2Te4 possesses excellent pulse modulation capabilities and holds significant potential as a promising ultrafast laser modulator for future research applications.
035504
You Li, Xiang Li, Mingliang Zhao and Ying Zhang
This paper presents a terahertz (THz) metasurface with polarization-frequency multiplexing capability that uses vanadium dioxide (VO2) to achieve dynamic wavefront control. The design leverages the phase transition of VO2 to switch between reflective and transmissive operational modes. In the metallic state, the metasurface operates reflectively to form four channels under linear polarizations (x-LP and y-LP) at 0.8 THz and 1.3 THz, generating diverse orbital angular momentum beams such as dual-, quadruple-, and octuple-vortex arrays. In the insulating state, it functions transmissively, creating two channels under x-LP and y-LP incidences at 0.8 THz via independent phase modulation, which produces multi-spot focal patterns like cross/X-shaped distributions and an eight-spot array. This work demonstrates full-space, six-channel (four reflective and two transmissive) wavefront manipulation, offering a viable strategy for developing high-performance devices for THz communication, imaging, and encryption systems.
035505
Wei-Ke Zou and Hong-Kang Zhao
The quantum dynamic behavior of photonic heat current through a circular mesoscopic Josephson junction device (CMJD) has been investigated under the perturbation of quantum electromagnetic field (QEMF). A quantum magnetic flux in squeezed coherent state and a static Aharonov–Bohm magnetic flux are threaded through the CMJD to adjust the output heat current periodically. Both of the phase and magnitude of QEMF contribute to the output heat current efficiently. Time-dependent source-drain photonic heat current is derived in temperature difference ΔT ≠ 0, and photon-pumping heat current is evaluated when ΔT = 0. The time-averaged heat conductance is achieved by expanding the time-averaged heat current around thermal equilibrium temperature. Two special types associated with the frequencies of QEMF and supercurrent in CMJD Ω0 and ω are considered especially in our numerical calculations. The applied QEMF induces the photonic heat current to form concrete package construction, and the modification behaviors of QEMF to the heat current is intimately dependent on the detailed combination of parameters. The photon-pumping heat current can be suppressed or enhanced by changing the squeeze parameter of QEMF at special combination of frequencies, and the sub-peaks of oscillation package can be split via increasing the squeeze parameter.
035506
Ruolin Wang, Yuanxi Cao, Maomao Zhang, Xiangqing Wang, Changchun Yang and Yong Zhu
Based on the dual-resonance composite right-/left-handed (CRLH) resonator, a design of dual-band microwave metamaterial filter aided by artificial intelligence (AI) algorithms optimizer is demonstrated in this paper. Specified by dual-mode electromagnetic (EM) resonant behavior, the emphatically illustrated metamaterial resonator is distinguished with characteristics of manipulated modes and couplings. Applied to the design of third-order topology dual-band filter, the circuit architecture consisting of multiple parameters requires efficient and accurate optimization for desired filtering performance. Then, an AI methodology integrating the co-simulator of HFSS-MATLAB with grey wolf algorithm (GWA) optimizer is employed to obtain the pre-set dua-band frequency response. With final filter layout exported, the prototype sample was fabricated. The implemented microwave properties measurement validates a good agreement of device performance between theoretical analysis and practical experiment that reveals the feasibility of described physical concept.
035507
Fuwei Wang, Wei Jiang, Xiangguo Zhai and Yang Gao
This work investigates how laser pretreatment parameters—scanning speed, spacing, and scan number—govern the microstructure and performance of electrodeposited Ni-Co-YSZ composite coatings. SEM with 3D profiling, XRD, microhardness, and Corrosion resistance were employed. The composite coatings mainly exhibited an FCC phase structure under the investigated processing conditions, dominated by a close-packed (111) texture. Moderate pretreatment markedly improves surface leveling/densification and refines crystallites, giving the smallest average size. The parameter set ‘400 mm s−1, 50 μm, five scans’ delivers the best overall properties: microhardness of 546.58 HV, corrosion current density of 6.309 μA cm−2, and polarization resistance of 6.929 kΩ cm2. The study found that the improved hardness and corrosion resistance of the coating are the result of a synergistic effect of Hall-Petch strengthening and the dense passivation film formed by the (111) texture of the coating. Insufficient or excessive laser treatment can lead to surface growth or remelting-induced coarsening and more defects on the substrate, thereby affecting the subsequent electrodeposition of the composite coating and reducing the hardness and corrosion resistance of the coating.
035508
Abdellah Ahlane, Zoubir Hricha and Abdelmajid Belafhal
The propagation behavior of an Elegant Hermite High-Order Cosine-hyperbolic Gaussian beam (EHHOChGB) in a strongly nonlocal nonlinear medium (SNNM) is theoretically investigated. Using the Snyder–Mitchell model in combination with the Huygens–Fresnel integral (Collins form), analytical expressions for the complex amplitude, rms beam width, and curvature radius are derived, enabling a detailed description of beam evolution in both weakly and strongly decentered configurations. The moment method is employed to evaluate the rms beam width and determine the critical power for self-trapping, providing quantitative insight into the transition between diffraction-dominated, solitonic, and over-focusing regimes. Numerical simulations reveal that the EHHOChGB exhibits periodic self-reconstruction governed by the interplay between diffraction and nonlocal self-focusing. At the critical input power, the beam maintains an invariant rms width and curvature, forming a stationary spatial soliton; below and above this threshold, it displays bounded breathing oscillations. The results show that the beam order parity, modulation index, and decentering parameter jointly control the propagation regime: odd Hermite orders generate dark-centered solitons, while even orders form bright breathing modes. Increasing the decentering parameter enhances internal interference and modulation sensitivity, demonstrating that the nonlocal medium acts as a tunable optical potential for structured beams. This study extends the Snyder–Mitchell framework to complex Hermite–cosh–Gaussian families and provides new insights into nonlocal self-trapping, power-dependent beam shaping, and nonlinear optical confinement.
Condensed matter and materials physics
035901
Adhinarayan Naembin Ashok, Sanjam Bedi, Taha Ashraf Ali Shaikh, Jai Aadhithya Ramesh and Adarsh Ganesan
This paper presents the possibility for inertial imaging of spatially patterned annular mass distributions of a circular graphene nanodrum resonator. By placing two distinct analytes in concentric annular regions, we harness the vibrational mode-specific sensitivities of the nanodrum to estimate their respective mass densities. An analytical formulation based on the Rayleigh-Ritz principle is developed to relate radial mass loading to modal frequency shifts. Finite element simulations are performed in COMSOL Multiphysics to obtain the shifts in the resonance frequency of vibrational modes under varying geometrical configurations of annular rings. By processing these frequency shifts through a transformation matrix, we estimate the concomitant mass distributions of annular rings. The results indicate that the estimation errors are lower for analytes placed near the antinodal regions of the dominant vibration mode, with the lowest error being 1.82% and 2.03% for the inner and outer annular regions, respectively. Furthermore, thinner annular rings demonstrate enhanced detection accuracy due to reduced modal overlap. This study demonstrates an analytical strategy for mass detection using a graphene nanodrum by providing insights into optimal analyte placement and structural design for high-precision multi-target mass sensing applications.
035902
A Ettouzani, A Zaghrane, G Merzouki, A Abbassi, B Manaut and S Taj
Thallium-based scintillators remain benchmarks for radiation detection, yet new materials with enhanced stability and performance are needed. Here, we employ density functional theory (DFT) and time-dependent DFT (TDDFT) to investigate the structural, electronic, and optical properties of orthorhombic ATlCl3 (A = Mg, Ca, Sr, Ba) perovskites. Formation energy analysis confirms their thermodynamic stability, with CaTlCl3 and SrTlCl3 dynamically stable, MgTlCl3 slightly unstable, and BaTlCl3 showing soft phonon modes. Elastic constants reveal systematic softening from Mg to Ba, though all satisfy Born stability criteria. Wide band gaps (4.58–5.28 eV) and high transparency support their suitability as scintillators. Predicted light yields range from 75, 763 to 87, 425 photons/MeV, with MgTlCl3 performing best. TDDFT calculations yield ultrafast radiative lifetimes (0.65–1.16 ns), highlighting their potential for fast-timing applications. These results position ATlCl3 perovskites as promising next generation scintillators, linking atomic scale properties to high performance radiation detection.
035903
M C González, H Martínez, J E Mastache-Mastache, V H Castrejón-Sánchez, P G Reyes and A Gómez
This study introduces a hybrid computational framework combining Monte Carlo simulations with partial differential equations (PDEs) to model gallium nitride (GaN) quantum dot (QD) formation during plasma-enhanced chemical vapor deposition (PECVD). The Monte Carlo approach captures stochastic particle deposition and nucleation on a 100 × 100 nm2 grid, incorporating energy-dependent sticking probabilities, electrostatic repulsion, and a log-normal QD size distribution (mean 4 nm, σ = 0.2). The PDE model describes surface diffusion and nucleation dynamics, linking stochastic outcomes to deterministic physical mechanisms. Experimental parameters include a deposition rate of 0.05 particles/(s·nm2), substrate temperature of 700 K, plasma pressure of 0.2–1.0 Torr, particle energy of 30 eV, and plasma temperature of 30,000 K. The simulations generate spatial distribution maps and size histograms, achieving QD densities of 3.0 × 1010 to 1.2 × 1011 cm−2, validated against experimental data for LED applications. Results show that increasing plasma pressure enhances QD density but reduces spatial uniformity due to faster nucleation outpacing diffusion, with implications for optimizing optoelectronic device performance. This framework provides a robust tool for tailoring deposition conditions to achieve high-density, uniform GaN QD arrays for advanced optoelectronic applications.
035904
Lijie Chen, Xiao Wang, Qingyang Deng, Yu Cheng and He Wang
Inspired by the multicellular fractal architecture of plant stems, this study proposes a novel biomimetic plant stem multi-cellular carbon fiber thin-walled tube (BPSMT). The structure integrates concentric nested carbon fiber tubes filled with polyurethane foam to mimic matrix-bound cellular hierarchies. Quasi-static axial compression tests reveal the progressive crushing failure modes and synergistic energy absorption exceeding the linear superposition of individual components. The arrangement of small tubes influences performance, which configurations with relative distribution outperform those with adjacent distribution in energy absorption. Increasing the small tube count enhances peak crushing force, with the 8-tube adjacent-distribution achieving the highest peak crushing force of 93.68 kN. The optimal relative distribution with 8 tubes demonstrates specific energy absorption of 59.61 kJ/kg, exceeding single tubes by 17%. The error of the crushing force prediction model is less than 6%. Compared to traditional foam-filled single tubes, BPSMTs have a 5.7% reduction in density and a 17% increase in specific energy absorption, confirming their superior lightweight and energy absorption capabilities.
035905
Aminul Islam, Ankur Gupta, Jency Rubia J, Yogesh Tripathi, Saptarshi Dutta, Lakshmi Narayanan V, Santosha Kumar Dwivedy and Ravindra Kumar Jha
The growing demand for graphene and its composites in electronics, energy storage, and structural engineering emphasizes the importance of scalable production strategies. In this work, we present a laser-assisted spray technique for integrating graphene onto polyamide surfaces, with loadings varying between 1 and 3 wt.%. During the deposition, graphene nanosheets infused and settled as solid building units, while a photo-curable resin served as a catalytic template for layered growth. The simultaneous laser exposure and spraying yielded graphene with high crystallinity, low defects (ID/IG = ~0.5), retaining a hexagonal carbon arrangement and uniform structural layers. The infused graphene enables significant enhancements in elastic modulus (∼2000 MPa) over pristine polyamide and notably enhanced in electrical conductivity, improved hydrophobicity (WCA=~74.5°), and superior corrosion resistance (Icorr= ~7.5 μA cm−2) etc The key improvements are attributed to the infusion of graphene, preserving its honeycomb lattice, adapting twisted architectures involved in multifunctional activities. Overall, these findings demonstrate the laser-assisted spray approach as an effective, scalable and cost-efficient fabricating route for flexible electronics.
035906
Hien Doan-Thi and Hai Pham-Van
We develop a symmetry-guided framework for phonon-driven enhancement of the hydrogen evolution reaction on two-dimensional catalysts. Using Chladni patterns and Noether’s theorem as organizing principles, we classify out-of-plane vibrations and map them to zone-center crystal symmetries, then apply site-symmetry projections to obtain a simple selection rule: under uniform coverage, only the fully symmetric breathing branch produces a coherent first-order shift of adsorption energetics of adsorption energetics, whereas other angular patterns cancel or are forbidden. Coupling this symmetry filter to a Brønsted–Evans–Polanyi kinetic description and a driven-oscillator response yields a closed-form rate enhancement and a practical figure of merit that combines linear coupling, modal stiffness, and mechanical quality factor, with microkinetic bandwidth entering as a low-pass constraint. We further introduce Chladni focusing, in which placing active regions along mode-selected annuli suppresses sign cancellations and amplifies the effective coupling. Case studies on hexagonal Mo2B2, Mo2C, and MoS2 illustrate mode selectivity, dual-site synergy, and clear operando fingerprints-resonant lineshapes, amplitude scaling, and symmetry null tests-that distinguish genuine mechanical acceleration from heating. The framework provides general, quantitative rules for selecting and driving lattice vibrations to surpass static Sabatier limits.
035907
Wen-Bo Yuan, Ting Yuan, Jie Shen, Chun-Ming Yang and Lei Hu
Facing the energy shortage and environmental pollution, the solar-cell application is becoming increasingly widespread. Single-layer (SL) In2Se3 is a famous two-dimensional (2D) semiconductor for strong visible absorption. However, its solar cell application has not been thoroughly studied, due to the lack of acceptor materials. Using first-principles calculations, this manuscript systematically investigates the structural, electronic, and power conversion efficiency (PCE) properties of In2S2Se/In2Se3 van der Waals heterojunction (vdWH), where SL In2S2Se acts as an acceptor material based on the work function. The electronic band structure reveals that In2S2Se/In2Se3 vdWH maintains a type-II band alignment under both biaxial and vertical strains, which is quite helpful for the carrier separation. In2S2Se/In2Se3 vdWH achieves a solar conversion efficiency of 16.52%, suggesting its great perspective as a solar cell material.
035908
Prabir Sarkar, Malay Kumar Das, Parameswara Rao Alapati, Banani Das, Anuj Upadhyay, Archna Sagdeo and Arvind Kumar Yogi
We present a surface-oriented synchrotron x-ray diffraction study of the smectic phases in N-(4-n-octyloxybenzylidene)-4′-butylaniline (8O.4) (I-SmA-SmB-SmF-CrG). The x-ray diffraction data from a planar-aligned sample were evaluated to extract orientational order parameters (OOP), layer thickness, and lateral molecular spacing across varying temperatures. The longitudinal and transverse correlation lengths were also determined from the diffraction patterns. Additionally the smectic translational order parameters and the tilt angles of the compound have been measured. The temperature-dependent birefringence (Δn) was measured to assess the orientational order parameter, which was subsequently compared with values derived from x-ray diffraction and mean field calculations. The homeotropically aligned sample yields an in-plane x-ray scattering profile, revealing significant hexatic bond-orientational order in the SmB, SmF and CrG phases. The diffraction patterns were examined to evaluate the bond orientational order (BOO). In the SmB phase, the BOO is approximately 0.85, with an in-plane correlation length of ∼107 Å, suggesting that this phase is a stacked hexatic structure with long-range bond orientational order but short-range positional order. The orthorhombic unit cell parameters for the SmB phase were also determined. Additionally, a new, more ordered smectic phase of this mesogen was identified as the smectic F phase, characterized by a C-centered monoclinic unit cell with axes a > b, where the unique axis b corresponds to the direction of molecular tilt. The technique used in this work will be helpful for future research on higher order smectic phases and other novel phases in liquid crystals.
035909
Jingcheng Yu, Jimin Fang, Jiaqi Zou, Tingyu Liu, Manzhuo Wang, Xiaoqiang Sun, Yuanda Wu and Daming Zhang
Dynamic manipulation of the magnitude and sign of nonreciprocal radiation is widely used in heat circulator, nonreciprocal solar cell, and nonreciprocal radiative cooling device. However, the effect of a magnetic field not being perfectly perpendicular to the incident plane on nonreciprocity remains unclear at present. To address this fundamental and critical issue, the effect of the magnetic field rotation angle on the nonreciprocal emitter with magnetophotonic crystal structures is comprehensively analyzed. The magnitude and sign of nonreciprocity can be dynamically manipulated via the rotation angle. The sign of nonreciprocity would be changed near the rotation angle of π/2. The devices exhibiting reciprocal behavior carry the rotation angle of π/2. Moreover, the nonreciprocal absorption and emission spectra can be dynamically manipulated by the rotation angle without additional introduction of graphene or phase change materials. The strong irreversible radiation is attributed to the optical Tamm plasmon polaritons at the interface between the photonic crystal and the metal. Coupled mode theory is adopted to elaborate the potential physical mechanism and further verify the validity of calculations. The generalized results that cover the case of multi-band nonreciprocal radiation prove the design flexibility and reliability. This work is believed to be helpful to improve the energy conversion efficiency of wavelength-selective thermophotovoltaic systems.
035910
Hao Wang, Hui Li and Lingna Meng
This study aims to develop curcumin-loaded polyvinylpyrrolidone (PVP)/zein-based degradable nanofibrous films for food packaging via centrifugal spinning. Five types of PVP/zein nanofibrous films with different curcumin loadings were prepared by centrifugal spinning, and characterized in terms of morphology, thermal stability, hydrophobicity, antioxidant activity, microbiological properties, and colorimetric indication. The results showed that the film with 160 mg curcumin exhibited the best overall antibacterial rate and antioxidant effect, with a water contact angle of 85.32° and a tensile strength of 5.71 MPa. The film also exhibits excellent reversible indicator functionality for volatile ammonia. Finally, the nanofibrous film was applied to fresh shrimp, and compared with the control group, the shelf life of shrimp was extended by three days, providing new insights and technological advancements for the development of next-generation intelligent food packaging materials.
035911
Sujuan Wang, Tao Shen, Chi Liu, Feifei Sun, Zuxiang Kou, Tianyu Sun and Delong Li
Decomposition gases (HF, CO, CF4, COF2) generated during partial discharge of the novel eco-friendly insulating gas perfluoroisobutyronitrile (C4F7N) pose a serious threat to the insulation performance of high-voltage insulating equipment, the safety of the natural environment and human health. Consequently, the development of a high-sensitivity gas-sensing material for leakage detection applications is of critical importance. In this study, the doping of MoTe2 with CuO clusters was investigated using density functional theory, and the adsorption behaviors of the pristine MoTe2 and the doped systems toward the four decomposition products were systematically compared. Theoretical calculations indicate that the doping of CuO clusters significantly enhances the interaction between the material and the gas molecules. Among the modified systems, CuO-MoTe2 exhibits the best adsorption performance for HF (−0.787 eV), while (CuO)2-MoTe2 shows the strongest adsorption for CO (−1.360 eV) and demonstrates a high sensing response value (104.61) and trace detection potential. Furthermore, the above adsorption systems demonstrate rapid recovery times, indicating favorable desorption performance and high reusability. Environmental interference studies reveal that air composition exerts a weak effect on adsorption, but the presence of humidity significantly enhances adsorption capacity. This work provides theoretical support for the application of (CuO)n-MoTe2 materials in detecting fault diagnostic gases in power equipment.
035912
Yusuph Mathias, V Kanaka Rao P, Joyce Elisadiki and Benard S Mwankemwa
Stannite-type quaternary crystals have emerged as promising, low-toxicity candidates for photocatalytic water treatment; however, the effectiveness of many conventional photocatalysts is often impeded by their large bandgaps when exposed to visible light. Therefore, there is a need to develop narrow-bandgap materials to maximize photocatalytic activity in the visible spectrum. This study employs first-principles calculations grounded in density functional theory (DFT) to examine the structural, electronic, and optical properties of magnesium (Mg)-doped Ag2(Zn)SnS4. A sampling mesh of 4 × 4 × 2 k-points was used for the calculations with a cut-off energy of 523 eV and the total energy convergence within 1 × 10−6 eV atom−1. The findings reveal that Mg-doping significantly reduces the direct bandgap from 1.153 eV (undoped) to a range of 0.597–0.688 eV, depending on the substitution ratio. This narrowing results in a considerable red shift in optical absorption from the ultraviolet to the visible spectrum. Moreover, Mg-doped structures demonstrate decreased reflectivity and improved conductivity in the visible spectrum compared to their undoped counterparts. These findings reveal that Mg substitution doping effectively enhances the semiconductor properties of Ag2(Zn)SnS4, making it a highly effective material for visible-light-driven photocatalytic water treatment applications.
035913
Shang Zhou, Jiangting Li, Xixu Bao, Yaqi Cheng, Xuemeng Zhang, Mingyue Zhang, Yifei Yang, Pei Chen, Yipeng Chen, Feng Ouyang et al
Understanding the influence and regulation mechanism of GaN substrate doping and hexagonal boron nitride (h-BN) layer thickness on the interfacial and electrical properties of single-layer graphene (SLG)/GaN heterojunctions is crucial for enhancing photoelectric performance of heterojunction devices. Here, we report the influence and regulation mechanism of the thickness of the h-BN insertion layer (1–5 layers) and the doping type of GaN (n- or p-type) on the electrical and interfacial physical properties of heterojunctions, including charge transfer, barrier height, and built-in potential. Our research shows that h-BN enhances barrier heights by suppressing charge transfer. Compared with the SLG/h-BN/n-GaN heterojunction with a larger depletion region area, the SLG/h-BN/p-GaN heterojunction with a smaller depletion region area has a stronger built-in electric field intensity. As h-BN thickness increases, the Fermi level of the SLG on both n-GaN and p-GaN substrates exhibits a gradual trend toward approaching the Dirac point. Furthermore, our calculations further reveal that the contact potential difference of the SLG/h-BN/n-GaN junction increases with h-BN thickness, while showing an opposite decreasing trend for the SLG/h-BN/p-GaN junction. Quantitative analysis of the interface trap charge density indicates that the surface characteristics of GaN can significantly influence the charge transport and interfacial physical properties at the heterojunction interface. These results will benefit recent topical application research on regulation strategies for 2D/3D heterojunction interfaces.
035914
Roopa, Bipul Kumar Pradhan, Dhruvika Tyagi, Sunil Singh Kushvaha and Senthil Kumar Muthusamy
The escalating emission of NO2 underscores the urgent need for effective gas sensors. Among metal oxides, tungsten oxide (WO3) is a promising candidate for NO2 detection. In this work, WO3 thin films were fabricated via thermal oxidation of DC magnetron-sputtered tungsten, with tungsten thickness varied from 50–200 nm. Grazing incidence x-ray diffraction (GIXRD) analysis confirmed crystallization and crystallite size enhancement with increasing thickness. Field emission scanning electron microscopy (FESEM) images depicted grain-based film structures, with surface Porosity increasing with thickness. Raman spectroscopy confirmed the chemical composition of WO3, while x-ray photoelectron spectroscopy (XPS) showed higher surface oxygen for the 100 nm film. The study explored thickness-dependent NO2 sensing at different operating temperatures. The study found that 100 nm film operated at 100 °C exhibited a high response of 1122% and 55% at room temperature, with lowest detection concentration of 100 ppb, which is highest compared to other thicknesses. Selectivity study found that the sensor is strongly selective towards NO2, along with promising repeatability.
035915
Hai Dang Ngo, Thi Kim Hang Pham, Bao Quan Tran, Khac Binh Nguyen, Minh Tuan Dao, An Hoang-Thuy Nguyen and Hoai Phuong Pham
This study examines the impact of substrate temperature on In-doped ZnO thin films were created on glass substrates via the RF-magnetron reactive sputtering method. The results indicate that the thin films possess a hexagonal Wurtzite structure, preferentially oriented along the (002) plane, which is perpendicular to the substrate surface. The crystallite sizes increases from 32.10 nm to 35.24 nm with an increase in substrate temperature. The UV–Vis transmission spectrum indicates that the films have an average transmittance over 85% in the visible light spectrum. The optical bandgap energy was measured via the Tauc method, ranging from 3.44 eV to 3.52 eV. Hall effect experiments indicate that the films exhibit low resistivity (∼10−4 Ω·cm), high carrier concentration exceeding 1021 cm−3, and electron mobility ranging from 3.03 to 6.72 cm2·V−1·s−1. The figure of Merit varies from 1.17 × 10−2 to 2.85 × 10−2 Ω−1, with the sample deposited at 250 °C achieving the maximum Seebeck coefficient (40.25 μV·K−1) and the ideal power factor (253.13 μW·m−1·K−2). These findings support the application of IZO thin films in optoelectronic and thermoelectric applications, with the goal of developing advanced technology based on environmentally friendly oxide materials.
Machine learning with application to physical science
036001
Ziwei Xiao and Junfeng Shen
Aspect category sentiment analysis (ACSA) involves identifying sentiment polarities of aspect categories explicitly or implicitly mentioned in sentences. However, existing methodologies continue to face challenges of data scarcity and insufficient contextual understanding in short text sentiment analysis, with notable accuracy decline when processing implicit aspect expressions. This paper proposes a prompt-based RoBERTa-semantic model (PRSM) to address these complex issues. PRSM transforms the ACSA task into a masked prediction problem through customized prompt learning templates, then leverages RoBERTa to extract contextual features and predict aspect categories. The introduction of a BiLSTM-based feature fusion structure effectively alleviates the problem of insufficient semantic representation, thereby enhancing the capture of multi-granular semantic features and deep semantic dependencies. Furthermore, a T5 autoregressive model generates explanatory text, compensating for inadequate understanding of implicit information. Experimental results across four benchmark datasets demonstrate that the proposed model significantly outperforms baseline approaches.
036002
Xingyue Tong, Shuai Ma, Yang Liu, Jiatong Sun, Xiaojin Cheng and Jianhua Shang
To overcome the limitations of traditional acoustic methods in sensitivity, classification accuracy, and deploy ability, this study proposes a novel framework that combines dual-comb multi-heterodyne interferometry (DCMHI) with deep learning. The system enables high-speed, high-sensitivity detection of sound pressure variations across frequencies using optical frequency combs (OFCs). A Mel-spectrogram dataset based on real marine mammal vocalizations was constructed, with multiple audio augmentation techniques employed to enhance data diversity and model generalization. Meanwhile, a Lightweight Marine Mammal Classification Network (LMMC-Net) model was designed, which integrates the Spatial-Channel Semantic-Guided Block (SCSGB) and the Linear Manhattan Self-Attention Block (LMASAB) to enable efficient acoustic feature extraction and accurate classification. Experiments show that the system achieves a minimum detectable sound pressure level (MDSPL) of 97.68 dB re 1 μPa for a 6 kHz pulse when the output signal-to-noise ratio (SNR) reaches 15 dB. Data augmentation improves classification accuracy by 2.6%. On the Mammalian Mel-Spectrogram Dataset (MMS-31 dataset), LMMC-Net outperforms YOLOv12 by 1.1% in Top-1 accuracy and 21.63% in inference speed, while reducing parameters by 41.62%, demonstrating strong potential for edge deployment.
036003
Zihe Cao, Huan Wang, Changli Shi, Yi Yang, Xiaolu Chen, Wenming Zhang, Zhangqi Yu, Qiang Lu and Bing Zhang
Predicting the band gap of titanium dioxide (TiO2) is crucial for its photocatalytic applications, yet first-principles calculations of band gap properties are often computationally expensive. Machine learning offers an efficient alternative, but predictive accuracy strongly depends on how structural features are represented. In this work, we investigate the structure–property relationships governing the TiO2 band gap by comparing different feature construction strategies. Traditional structure descriptors based on local atomic environments show limitations in effectively encoding complex structural correlations. In contrast, graph neural network (GNN) models more effectively learn structure–property relationships and achieve higher prediction accuracy. Further deconstruction of the model reveals that its predictive performance relies on a synergistic integration of chemical identity, local geometry, and non-local structural information. Feature visualization further shows that the model learns task-adaptive representations. These findings highlight the necessity of incorporating non-local structural information for predicting global properties and provide new insights into the machine learning driven materials design.
036004
Abdullah Al-Yaari, Youjun Deng, Muhammad Abdullahi and Muhammad Shakeel
We propose NU-SSL GAN, a Neuroevolutionary semi-supervised generative adversarial network designed for effective learning under extreme label scarcity. We evaluate NU-SSL GAN on two challenging medical image classification tasks: the HAM10000 skin lesion dataset (pigmented lesion images) and the BreCaHAD breast cancer histopathology dataset (microscopic biopsy images). Our results show that NU-SSL GAN achieves the highest classification accuracy among the GAN-based SSL methods tested, reaching 88.4% on HAM10000 and 98.5% on BreCaHAD under a 7% labeled regime. in both classification accuracy and image generation quality, and attains the highest classification accuracy on both datasets while producing realistic and diverse synthetic images with only a tiny fraction of labeled examples. Notably, NU-SSL GAN markedly narrows the gap with fully supervised performance despite using far fewer labels and training epochs. We attribute NU-SSL GAN’s success to its dynamic evolution of generator and discriminator architectures, which foster robust learning dynamics, avoid mode collapse, and ensure stable convergence under severe label constraints.
036005
The following article is Open access
Manisha Khulbe and Aryan Singhal
Quantum physics is used for communications, computing, and sensor applications beyond classical capabilities. Uniform Circular Array (UCA) antennas are one of the means of generating Orbital Angular Momentum waves (OAM), which are known as quantum waves that travel at microwave frequencies. The research focuses on the mathematical analysis of OAM wave generation using both classical and quantum optics. These quantum waves are simulated by developing cross-phased omnidirectional antennas and UCAs with 10-element patches using MATLAB. Antennas generating OAM waves are tested at frequencies 1.8 GHz, 40 GHz, 50 GHz, and 90 GHz. The antenna radiation range, elevation angle, and azimuth angles are analyzed. One UCA Antenna at 6 GHz is designed, fabricated, and tested for its radiation, circular polarization, and gain. UCAs are used in applications such as estimating radar cross-section (RCS) and locating the direction of targets. Simulations are carried out to detect targets in 3D (R, θ, Φ) and to calculate the RCS. Since deep learning (DL) techniques help solve large databases of RCS calculations, we utilise gradient descent and backpropagation methods for target detection. These techniques can estimate the position and surface structure of the target using RCS. DL models prove their greater robustness in direction of arrival estimation (DoA) in low SNR conditions. It shows high accuracy and less computational complexity, even in nonlinearity, as compared to traditional optimization methods such as the MUSIC algorithm. Hence, the design of Quantum antennas in DL is proposed for future wireless communication, frequency-aware designs, and high-resolution sensing.
036006
S Dahal, B Guragain, S K Yadav and N P Adhikari
Elastic constants govern how crystalline solids respond to stress, resist deformation, and transmit elastic waves. For cubic crystals, the mechanical behavior can be fully described using three independent elastic constants,
,
and
. These constants can be determined experimentally or through computational approaches such as molecular dynamics or density functional theory (DFT). However, experimental measurements are often challenging, while DFT calculations, though accurate, are computationally expensive. To address this, we have used a machine learning (ML) framework to predict and classify the elastic properties of ternary cubic alloys. A multi-output XGBoost model was trained on Matminer-featurized data from the Materials Project and validated against DFT calculations for 12 half-Heusler alloys, showing excellent agreement. We then applied the model to predict the elastic constants of 972 ternary materials, finding that 970 are mechanically stable and exhibit strong internal correlations among their elastic properties. Additionally, we employed unsupervised learning via K-means clustering to categorize these materials into four mechanical classes such as brittle and soft, ductile with moderate stiffness, stiff and strong, and highly ductile and soft. This integrated approach establishes a high-throughput pathway for accelerating the discovery and targeted design of materials with tailored mechanical performance.
Reply
Comment and Reply
037001
Yue Qin, Yuqin Yao and Ye-Hui Huang
Retraction
039701
The following article is Free article
M P Kedarnath, N Ramasamy, M Dev Anand and J Bensam Raj