A research team led by Prof. Gao Xiaoming from the Hefei Institutes of Physical Science of the Chinese Academy of Sciences has improved residual neural networks to accurately classify and identify microplastics using low-quality Raman spectra, even under non-ideal experimental conditions.
“It detects and classifies microplastics when the data is cluttered with noise,” said Prof. Gao, “and it does this without overloading computing power.”
The research results are published in Talanta.
Guiding light around dynamic regions of a scattering object by means of propagating light through the most ‘stable’ channel within a moving scattering medium is demonstrated, potentially advancing fields such as deep imaging in living biological tissue and optical communications through turbulent air and underwater.
Magnetic materials have become indispensable to various technologies that support our modern society, such as data storage devices, electric motors, and magnetic sensors.
High-magnetization ferromagnets are especially important for the development of next-generation spintronics, sensors, and high-density data storage technologies. Among these materials, the iron-cobalt (Fe-Co) alloy is widely used due to its strong magnetic properties. However, there is a limit to how much their performance can be improved, necessitating a new approach.
Some earlier studies have shown that epitaxially grown films made up of Fe-Co alloys doped with heavier elements exhibit remarkably high magnetization. Moreover, recent advances in computational techniques, such as the integration of machine learning with ab initio calculations, have significantly accelerated the search for new material compositions.
In a new publication in Nature Materials, an international team of researchers has developed groundbreaking artificial chains of the iconic “olympicene” molecules to realize the antiferromagnetic (AF) spin-½ Heisenberg model, a flagship quantum spin model that has been the cornerstone of quantum magnetism, since the seminal work of Bethe, for almost a century now. This study makes nanographenes (NGs) an ideal platform for realizing and studying highly entangled quantum spin systems, with potential applications in insulator-based AF spintronics.
In one-dimensional quantum magnets, strong quantum fluctuations prevent spontaneous symmetry breaking, leading to the formation of quantum-disordered many-body states such as resonating valence bond states. Half-integer spin chains are expected to exhibit a gapless excitation spectrum in the thermal dynamic limit, with the elemental excitations comprising at least two fractional spin-½ with well-defined energy-momentum relation, known as spinons.
In finite length, confinement effects introduce a quantization gap, which gradually approaches zero as the chain length increases (L→∞). Despite the theoretical appeal, the experimental realization of the isotropic spin-½ Heisenberg model faces significant challenges. Furthermore, the lack of access to well-defined finite chains hampers systematic studies on how spin excitations evolve with chain length and how even-and odd-numbered chains exhibit distinct behaviors.
Have you ever left a bottle of liquid in the freezer, only to find it cracked or shattered? To save you from tedious freezer cleanups, researchers at the University of Amsterdam have investigated why this happens, and how to prevent it. They discovered that while the liquid is freezing, pockets of liquid can get trapped inside the ice. When these pockets eventually freeze, the sudden expansion creates extreme pressure—enough to break glass.
“Newton had an apple fall on his head. I found my freezer full of broken glass,” says Menno Demmenie, first author of the new study that was recently published in Scientific Reports.
He continues, “The usual explanation for frost damage is that water expands when it freezes, but this does not explain why half-filled bottles also burst in our freezers. Our work addresses how ice can break a bottle even when it has plenty of space to expand into.”
A symmetry violation has been observed in a particle-decay process that—together with five related decays—could shed light on the matter–antimatter imbalance in the Universe.
The known Universe has some 1012 galaxies that are made out of matter and no galaxies that are made out of antimatter. This is a surprising result because matter and antimatter are expected to exist in equal quantities. More formally, matter and antimatter are related by a symmetry known as CP symmetry, which states that a particle and its antiparticle should obey the same laws of nature. A necessary condition for the observed imbalance between matter and antimatter in the Universe is therefore a violation of CP symmetry—for a review see H. R. Quinn and Y. Nir [1]. Solving this puzzle has driven extensive experimental efforts that have revealed such a violation in different particle sectors. The Large Hadron Collider Beauty (LHCb) Collaboration at CERN has now measured a CP violation in a certain decay channel of B ±].
Altermagnets are arguably the hottest objects in magnetism right now (see Viewpoint: Altermagnetism Then and Now). Over the past year, researchers have delivered experimental evidence for this new type of magnet, but they have yet to harness the behavior for applications. Now three independent groups have proposed methods for electrically tuning the properties of altermagnets [1– 3]. If implemented, the findings could allow the use of altermagnets in next-generation spintronics devices.
Altermagnets can be thought of as a cross between antiferromagnets and ferromagnets. Like antiferromagnets the materials lack net magnetization—the magnetic spins of the atomic lattice are aligned in opposing directions. Like ferromagnets they have magnetically sensitive energy levels and display electronic band structures that are split into spin-up and spin-down bands. This splitting can be used to polarize an electronic current, as one spin state will flow through the material more easily. The combination of these properties could allow researchers to create spintronics devices that operate more rapidly and with greater efficiency than those currently in use, but for that, they first need a way to manipulate the spin properties of an altermagnet.
The proposed methods of the three teams (a group led by Tong Zhou of the Eastern Institute of Technology, Ningbo, China; Libor Šmejkal of the Max Planck Institute for the Physics of Complex Systems, Germany; and a group led by Qihang Liu of the Southern University of Science and Technology, China) all use electric fields for this switching. Controlling magnetism with electricity is particularly attractive because electric fields are much easier to manipulate and integrate into modern electronic devices than magnetic fields. Electrical tuning is potentially also faster (subnanosecond) and could use less energy, two crucial properties for the development of high-speed, low-power spintronic devices.
Illuminating a metasurface with a laser can enable the rapid modulation of the polarization of terahertz light transmitted through the metasurface.
Time crystals realized in the so-called quasiperiodic regime hold promise for future applications in quantum computing and sensing.
In ordinary crystals, atoms or molecules form a repeating pattern in space. By extension, in quantum systems known as time crystals, particles form a repeating pattern in both space and time. These exotic systems were predicted in 2012 and first demonstrated in 2016 (see Viewpoint: How to Create a Time Crystal). Now Chong Zu at Washington University in St. Louis and his colleagues have experimentally realized a new form of time crystal called a discrete-time quasicrystal [1]. The team suggests that such states could be useful for high-precision sensing and advanced signal processing.
Conventional time crystals are created by subjecting a collection of particles to an external driving force that is periodic in time. Zu and his colleagues instead selected a quasiperiodic drive in the form of a structured but nonrepeating sequence of microwave pulses. The researchers applied this quasiperiodic drive to an ensemble of strongly interacting spins associated with structural defects, known as nitrogen-vacancy centers, in diamond. They then tracked the resulting dynamics of these spins using a laser microscope.
BL Lacertae, an enigmatic blazar, has shattered long-held classification norms, leaving astronomers baffled. Originally mistaken for a variable star, this active galaxy emits high-energy jets that have suddenly defied expectations.
Observations from 2020–2023 revealed that BL Lacertae doesn’t neatly fit into any of the three known blazar categories, shifting unpredictably between classifications. This rapid transformation, particularly in X-ray emissions, has sparked intense debate about the underlying physics. Could it be an entirely new type of blazar? Or is an unknown mechanism at play, altering its radiation patterns at unprecedented speeds?
Mysterious Blazar Challenges Astronomers.