On-demand rotational manipulation of microparticles and zebrafish larvae with acoustics

Scientists have demonstrated an innovative method for manipulating the rotation of microparticles and small-scale living organisms, such as zebrafish larvae, using bulk acoustic waves (BAW) with orthogonal phases. This technique enables precise, on-demand control of rotational movement, opening new avenues for studying biological processes and manipulating materials at the microscale. The advance is based on the application of acoustic fields that generate torques on objects, allowing them to rotate in a specific plane.

Simulations Evaluate Production Routes of Cesium-128 for Nuclear Medicine

Researchers have utilized advanced computational simulations to explore the most efficient and viable routes for the production of the positron-emitting radionuclide cesium-128 (¹²⁸Cs). This isotope is of significant interest for applications in nuclear medicine, particularly in positron emission tomography (PET), due to its short half-life and emission characteristics. The study focused on evaluating various nuclear reactions induced by protons and deuterons on xenon (Xe) and iodine (I) targets. To conduct this investigation, the team employed a set of consolidated simulation tools in nuclear physics. GEANT4 codes were used for particle transport modeling and interaction with matter, while TALYS and EMPIRE codes were utilized for simulating nuclear reactions and calculating cross-sections. These simulations allowed for the prediction of ¹²⁸Cs production yields and the presence of radionuclidic impurities as a function of incident particle energy and target material. The primary goal was to identify optimal conditions that maximize the production of pure ¹²⁸Cs, minimizing the co-production of other isotopes that could interfere with PET imaging or increase patient radiation dose. The simulation results provided crucial data on the cross-sections of relevant nuclear reactions, indicating which projectile-target combinations are most promising. This computational modeling approach is fundamental for planning production experiments in cyclotrons and accelerators, reducing the need for costly and time-consuming empirical testing. Optimizing ¹²⁸Cs production could facilitate its availability for research and clinical use, opening new avenues for disease diagnosis and monitoring via PET.

Multiband terahertz metasurface for refractive index biosensing

Researchers have numerically proposed a multiband terahertz (THz) metasurface with a high quality (Q) factor for use as a biosensor. This design aims to improve the detection of biomolecules by measuring changes in the refractive index, a crucial technique in medical diagnostics and biological research. The novelty lies in its ability to operate at multiple THz frequencies, which could allow for more detailed characterization of samples. The study focuses on a metasurface design that exhibits high-Q resonances, meaning that the interactions between THz light and the sample are more intense and localized. This high Q factor is essential for detecting subtle changes in the refractive index of biomolecules. Numerical simulation has allowed for the optimization of the metasurface's geometry and materials to achieve these properties, overcoming some limitations of current THz biosensors, such as their sensitivity and ability to operate over a broad spectral range. The relevance of this work lies in the potential to develop more efficient and versatile THz biosensors. The multiband capability not only enhances detection specificity but also opens the door to simultaneous identification of multiple analytes or characterization of complex sample properties. Although this is a numerical study, it lays the groundwork for the fabrication and experimentation of these devices, which could lead to significant advances in early disease diagnosis and the understanding of biological processes at the molecular level.

Multilayer engineering enhances energy storage in ferroelectrics

Researchers have developed a novel multilayer engineering strategy to significantly improve the energy storage capacity in tungsten bronze-based ferroelectric materials. This breakthrough is crucial for the development of more efficient and compact electronic devices, especially in applications requiring high energy and power density, such as multilayer ceramic capacitors (MLCCs). The key to this improvement lies in the application of multiple dielectric layers, which allows for optimizing the electrical response of the material. Ferroelectric materials, known for their spontaneous polarization capability, are fundamental in energy storage. However, their efficiency is limited by factors such as hysteresis and dielectric losses. The new technique addresses these limitations by modifying the microstructure and interfacial properties of the material. The results show a remarkable increase in stored energy density compared to conventional ferroelectric materials. This approach not only enhances the performance of existing materials but also opens new avenues for the design of future energy storage devices. This technology is expected to have a significant impact on power electronics, renewable energy systems, and electric vehicles, where the demand for high-performance energy storage components is continuously growing.

Neural Network Solves High-Dimensional Sine-Gordon Equations

Researchers have developed a gradient-enhanced physics-informed neural network (PINN) with adaptive loss weighting to tackle high-dimensional non-linear Sine-Gordon problems. This novel approach allows for more accurate and stable solutions to complex partial differential equations (PDEs) that describe physical phenomena such as Josephson junctions, coupled pendulum chains, or optical pulse propagation in fibers. The Sine-Gordon equation is known for its non-linear nature and the emergence of soliton-type solutions, making it a computational challenge, especially in scenarios with multiple spatial variables. Traditional PINNs, which embed physical laws directly into the machine learning loss function, often struggle with convergence and accuracy in high-dimensional problems or those with complex dynamics. The introduced enhancement addresses these limitations by incorporating gradient information and dynamically adjusting the weight of different loss function terms during training, guiding the network towards more precise and stable solutions. The proposed methodology represents a significant advancement in the application of artificial intelligence to computational physics. By overcoming the inherent difficulties of high-dimensional Sine-Gordon equations, new avenues are opened for the simulation and analysis of complex physical systems that were previously intractable or required prohibitive computational power with traditional numerical methods. This could accelerate the design of superconducting devices or the understanding of non-linear wave phenomena.

Analysis of a novel conservative chaotic system and its application in cryptography

A recent study has explored the dynamics of a novel conservative chaotic system, characterized by its ability to preserve volume in phase space. These types of systems, unlike dissipative ones, do not lose energy over time and exhibit complex yet predictable long-term behavior, making them of interest for various applications, including information security. The research focuses on understanding the intrinsic properties of this new system and how its chaotic features can be harnessed.