Toggle light / dark theme

Imposing time-dependent strain on a magnetic disk induces vortex dynamics and offers a path toward energy-efficient spintronic devices.

Nanoscopic magnetic vortices made from electron spins could be used in spintronic computers (see Research News: 3D Magnetism Maps Reveal Exotic Topologies). To this end, researchers need an energy-efficient way to excite these vortices into a so-called gyrotropic mode—an orbital motion of the vortex core around the central point. The direction of this orbital motion would determine which of two binary states the vortex represents. Vadym Iurchuk at the Helmholtz-Zentrum Dresden-Rossendorf, Germany, and his colleagues have now demonstrated such a method by imposing a time-varying strain on a magnetic material [1].

The excitation of gyration dynamics by an oscillating strain was suggested by a separate team in 2015 [2]. The idea involves depositing a magnetic film, in which magnetic vortices form spontaneously, on a piezoelectric substrate. Applying an alternating voltage to the substrate transfers a time-varying mechanical strain to the film, dynamically perturbing its magnetic texture. This perturbation displaces a vortex core from its equilibrium position, thereby exciting the gyrotropic mode.

Scientists are finding ways to use quantum effects to create groundbreaking thermal devices that can help cool electronic systems. The quantum thermal transistor is one of the most exciting innovations in this field. While the current works surrounding this device are still theoretical, recent advancements in the fabrication of qubits using quantum dots and superconducting circuits have created a growing sense of optimism.

Scientists have discovered that cosmic filaments, the largest known structures in the universe, are rotating. These massive, twisting filaments of dark matter and galaxies stretch across hundreds of millions of light-years and play a crucial role in channeling matter to galaxy clusters. The finding challenges existing theories, as it was previously believed that rotation could not occur on such large scales. The research was confirmed through both computer simulations and real-world data, and it opens up new questions about how these giant structures acquire their spin.

After reading the article, a Reddit user named Kane gained more than 100 upvotes with this comment: “What if galaxy clusters are like neuron and glial clusters in a brain. And dark matter is basically the equivalent of a synapse. It connects galaxies and matter together and is responsible for sending quantum information back and forth like a signal chain.”

A team of microchip engineers at Pragmatic Semiconductor, working with a pair of colleagues from Harvard University and another from Qamcom, has developed a bendable, programmable, non-silicon 32-bit RISC-V microprocessor. Their research is published in the journal Nature.

Over the past several years, hardware manufacturers have been developing bendable microprocessors for use in . A bendable device with bendable components would allow for the creation of 24-hour sensors that could be applied to any part of the body.

For this new project, the research team developed an inexpensive circuit board that could be bent around virtually any curved object. The material was made using indium gallium zinc oxide instead of the more rigid silicon.

Masafumi Oizumi: Unsupervised alignment of qualia structures: Towards direct communication of experience.

Pre-ASSC (2024, June 30, Sun) Satellite Workshop Registration Form: Structural approaches to consciousness: Qualia Structure and Integrated Information Theory.

— program -
Day: June 30, 2024 (Sun)

Time: 9:00–17:00
Place: Ito Hall, Ito International Research Center, University of Tokyo.

Aims:
In this satellite symposium, hosted by the Qualia Structure, we will discuss the current state of the structural approaches to consciousness from empirical, computational and theoretical perspectives. The satellite will be open to any researcher (but registration is required) who are interested in the structural approach to consciousness. A one day event will be a mixture of lectures and poster presentations.

Registration: Free.

The foundation of this simulation, as described by the team, is a well-known cosmological model that describes the universe as expanding uniformly over time. The researchers modeled how a quantum field, initially in a vacuum state (meaning no particles are present), responds to this expansion. As spacetime stretches, the field’s oscillations mix in a process that can create particles where none previously existed. This phenomenon is captured by a transformation that relates the field’s behavior before and after the universe expands, showing how vibrations at different momenta become entangled, leading to particle creation.

To understand how many particles are generated, the researchers used a mathematical tool called the Bogoliubov transformation. This approach describes how the field’s vacuum state evolves into a state where particles can be detected. As the expansion rate increases, more particles are produced, aligning with predictions from quantum field theory. By running this simulation on IBM quantum computers, the team was able to estimate the number of particles created and observe how the quantum field behaves during the universe’s expansion, offering a new way to explore complex cosmological phenomena.

According to the team, the most notable result of the study was the ability to estimate the number of particles created as a function of the expansion rate of the universe. By running their quantum circuit on both simulators and IBM’s 127-qubit Eagle quantum processor, the researchers demonstrated that they could successfully simulate particle creation in a cosmological context. While the results were noisy—particularly for low expansion rates—the error mitigation techniques used helped bring the outcomes closer to theoretical predictions.

The physics of drop motion in microchannels is fundamental to provide insights when designing applications of drop-based microfluidics. In this paper, we develop a boundary-integral method to simulate the motion of drops in microchannels of finite depth with flat walls and fixed depth but otherwise arbitrary geometries. To reduce computational time, we use a moving frame that follows the droplet throughout its motion. We provide a full description of the method, including our channel-meshing algorithm, which is a combination of Monte Carlo techniques and Delaunay triangulation, and compare our results to infinite-depth simulations. For regular geometries of uniform cross section, the infinite-depth limit is approached slowly with increasing depth, though we show much faster convergence by scaling with maximum vs average velocities. For non-regular channel geometries, features such as different branch heights can affect drop partitioning, breaking the symmetric behavior usually observed in regular geometries. Moreover, non-regular geometries also present challenges when comparing the results for deep and infinite-depth channels. To probe inertial effects on drop motion, the full Navier–Stokes equations are first solved for the entire channel, and the tabulated solution is then used as a boundary condition at the moving-frame surface for the Stokes flow inside the moving frame. For moderate Reynolds numbers up to Re = 5, inertial effects on the undisturbed flow are small even for more complex geometries, suggesting that inertial contributions in this range are likely small. This work provides an important tool for the design and analysis of three-dimensional droplet-based microfluidic devices.