There is an important and unresolved tension in cosmology regarding the rate at which the universe is expanding, and resolving this could reveal new physics.  Astronomers constantly seek new ways to measure this expansion in case there may be unknown errors in data from conventional markers such as supernovae. Recently, researchers including those from the University of Tokyo measured the expansion of the universe using novel techniques and new data from the latest telescopes. Their method exploits the way light from extremely distant objects takes multiple pathways to get to us. Differences in these pathways help improve models on what happens at the largest cosmological scales, including expansion.

Future events such as the weather or satellite trajectories are computed in tiny time steps, so the computation must be both efficient and as accurate as possible at each step lest errors pile up. A Kobe University team now introduces a new method that uses deep learning for creating tailored, accurate simulations that respect physical laws, while also being more computationally efficient.

Every bit of technology fails eventually, whether time, wear-and-tear, adverse conditions, or perhaps a mixture of all three are the cause of its demise. Yet predicting when technology might fail is notoriously difficult. When designing equipment for clear-energy systems, engineers often over compensate for this by designing it to withstand more stress than needed. This generates waste, and can sometimes inadvertently increase the wear and tear on materials. Now, an international research team has developed a way to more accurately predict how long mechanical-equipment used in green technology will last, potentially leading to more efficient design processes.

Silver-based atomic switches that create stable electrical connections between individual molecules and electrodes have been developed by researchers from Japan, addressing a key challenge in wiring molecular electronics. The switch operates by forming and breaking silver atomic filaments when a voltage is applied and reversed, corresponding to the “on” and “off” states. This method enables the scalable integration of molecular components, paving the way for ultra-compact and energy-efficient circuits built from single molecules.

Smart polymers change their properties in response to temperature, stress, or other stimuli, making them useful in drug delivery and soft robotics. But a major hurdle has been understanding how they behave when flowing or being stretched—conditions they face in real-world use. Now, researchers from Tokyo University of Science have developed a custom rheo-impedance device that provides the first look at these details, paving the way for more reliable and responsive smart materials.

A recent study explores how Japanese learners of French produce ambiguous speech errors. Using a specialized assessment tool, researchers found that many mistakes once considered purely “phonetic” may actually be rooted in morpholexical misunderstanding, and vice-versa. The findings offer teachers valuable guidance for improving pronunciation pedagogy and error correction. This work opens new possibilities for assessing ambiguity in second-language speech and better understanding the interplay between sound and meaning in learning French.

Magnetic resonance imaging (MRI) is invaluable in the medical world. But despite all the good it does, there is room for improvement. One way to enhance the sensitivity of MRI is called dynamic nuclear polarization (DNP), where target molecules for imaging are modified so they form clearer images when scanned with an MRI machine. But this technique requires some special crystalline materials mixed with polarizing agents which are difficult to create. For the first time, researchers including those from the University of Tokyo demonstrate the use of molecules called fullerenes as polarizing agents. Their new method can make DNP targets sufficient to yield far greater clarity when imaged with an MRI machine, with potential benefits in various medical applications.

Kyoto, Japan --  "Why are we here?" is humanity's most fundamental and persistent question. Tracing the origins of the elements is a direct attempt to answer this at its deepest level. We know many elements are created inside stars and supernovae, which then cast them out into the universe, yet the origins of some key elements has remained a mystery.

Chlorine and potassium, both odd-Z elements -- possessing an odd number of protons -- are essential to life and planet formation. According to current theoretical models, stars produce only about one-tenth the amount of these elements observed in the universe, a discrepancy that has long puzzled astrophysicists.

This inspired a group of researchers at Kyoto University and Meiji University to examine supernova remnants for traces of these elements. Using XRISM -- short for X-Ray Imaging and Spectroscopy Mission, an X-ray satellite launched by JAXA in 2023 -- the team was able to perform high-resolution X-ray spectroscopic observations of the Cassiopeia A supernova remnant within the Milky Way.

A research team led by Ryo Shimano of the University of Tokyo has successfully visualized two distinct mechanisms through which up and down spins, inherent properties of electrons, switch in an antiferromagnet, a material in which spin alignments cancel each other out. One of the visualized mechanisms provides a working principle for developing ultrafast, non-volatile magnetic memory and logic devices, which could be much faster than today’s technologies. The findings are published in the journal Nature Materials.

A novel treatment (MA-5) developed by a Tohoku University-led research team will soon undergo phase II clinical trials at four medical institutions across Japan. MA-5 could provide hope for patients with mitochondrial diseases – as there are currently no approved treatments available.