Toggle light / dark theme

A machine learning predictor enhances capability for solving intricate physical problems

In a recent development at Fudan University, a team of applied mathematicians and AI scientists has unveiled a cutting-edge machine learning framework designed to revolutionize the understanding and prediction of Hamiltonian systems. The paper is published in the journal Physical Review Research.

Named the Hamiltonian Neural Koopman Operator (HNKO), this innovative framework integrates principles of mathematical physics to reconstruct and predict Hamiltonian systems of extremely-high dimension using noisy or partially-observed data.

The HNKO framework, equipped with a unitary Koopman structure, has the remarkable ability to discover new conservation laws solely from observational data. This capability addresses a significant challenge in accurately predicting dynamics in the presence of noise perturbations, marking a major breakthrough in the field of Hamiltonian mechanics.

Physicists Discover Evidence of Time Being Reversible in Glass

Time’s inexorable march might well wait for no one, but a new experiment by researchers at the Technical University of Darmstadt in Germany and Roskilde University in Denmark shows how in some materials it might occasionally shuffle.

An investigation into the way substances like glass age has uncovered the first physical evidence of a material-based measure of time being reversible.

For the most part the laws of physics care little about time’s arrow. Flip an equation describing the movement of an object and you can easily calculate where it started. We describe such laws as time reversible.

Physicists Develop New Significantly More Efficient Solar Cell

Physicists at Paderborn University have enhanced solar cell efficiency significantly using tetracene, an organic material, based on complex computer simulations. They discovered that defects at the tetracene-silicon interface boost energy transfer, promising a new solar cell design with drastically improved performance.

Physicists at Paderborn University have used complex computer simulations to create a novel solar cell design that boasts substantially higher efficiency than existing options. The enhancement in performance is attributed to a slender coating of an organic compound named tetracene. The results have recently been published in the renowned journal Physical Review Letters.

“The annual energy of solar radiation on Earth amounts to over one trillion kilowatt-hours and thus exceeds the global energy demand by more than 5,000 times. Photovoltaics, i.e. the generation of electricity from sunlight, therefore offers a large and still largely untapped potential for the supply of clean and renewable energy. Silicon solar cells used for this purpose currently dominate the market, but have efficiency limits,” explains Prof Dr Wolf Gero Schmidt, physicist and Dean of the Faculty of Natural Sciences at Paderborn University. One reason for this is that some of the energy from short-wave radiation is not converted into electricity, but into unwanted heat.

Supernova Scavenger Hunt: Cracking the Case of Cosmic Ghost Stars

Supernovae–stellar explosions as bright as an entire galaxy–have fascinated us since time immemorial. Yet, there are more hydrogen-poor supernovae than astrophysicists can explain. Now, a new Assistant Professor at the Institute of Science and Technology Austria (ISTA) has played a pivotal role in identifying the missing precursor star population. The results, now published in Science, go back to a conversation the involved professors had many years ago as junior scientists.

The Enigma of Hydrogen-Poor Supernovae

Some stars do not simply die down, but explode in a stellar blast that could outshine entire galaxies. These cosmic phenomena, called supernovae, spread light, elements, energy, and radiation in space and send galactic shock waves that could compress gas clouds and generate new stars. In other words, supernovae shape our universe. Among these, hydrogen-poor supernovae from exploding massive stars have long puzzled astrophysicists. The reason: scientists have not been able to put their finger on their precursor stars. It is almost as if these supernovae appeared out of nowhere.

A new chapter for all-attosecond spectroscopy: Researchers achieve 1 kilohertz repetition rate

A team of researchers from the Max Born Institute in Berlin has, for the first time, demonstrated attosecond-pump attosecond-probe spectroscopy (APAPS) at a repetition rate of 1 kilohertz. This became possible by the development of a compact, intense attosecond source using an out-of-focus generation geometry. The approach opens new avenues for the investigation of extremely fast electron dynamics in the attosecond regime.

The first generation of pulses (1 attosecond corresponds to 10-18 seconds) at the turn of this century has enabled unprecedented insights into the world of electrons. For their pioneering work, first leading to the demonstration of attosecond pulses in 2001, Anne L’Huillier, Pierre Agostini, and Ferenc Krausz were awarded the Nobel Prize in Physics in 2023.

Current attosecond techniques, however, come with an important drawback: To be able to record a movie in a pump-probe experiment, an attosecond pulse typically has to be combined with a femtosecond pulse (1 femtosecond corresponds to 10-15 seconds) whose optical cycles (a few femtoseconds long) is used as a clock with attosecond resolution. This constitutes a limitation for the investigation of electron dynamics on attosecond timescales.

Photon upconversion: Steering light with supercritical coupling

National University of Singapore researchers and their collaborators have unveiled a novel concept termed “supercritical coupling” that enables a several-fold increase in photon upconversion efficiency. This discovery not only challenges existing paradigms, but also opens a new direction in the control of light emission.

Photon upconversion, the process of converting low-energy photons into higher-energy ones, is a crucial technique with broad applications, ranging from super-resolution imaging to advanced photonic devices. Despite considerable progress, the quest for efficient upconversion has faced challenges due to inherent limitations in the irradiance of lanthanide-doped nanoparticles and the critical coupling conditions of optical resonances.

The concept of “supercritical coupling” plays a pivotal role in addressing these challenges. This fundamentally new approach, proposed by a research team led by Professor Liu Xiaogang from the Department of Chemistry, NUS and his collaborator, Dr. Gianluigi Zito from the National Research Council of Italy leverages on the physics of “bound states in the continuum” (BICs).

/* */