Despite a surge of interest in Europe in establishing autonomy in space systems, there remains skepticism that one of the biggest efforts along those lines.
China has launched its first mission to retrieve samples from an asteroid. The Tianwen-2 mission launched at 01:31 a.m. local time on 28 May from the Xichang satellite launch center, southwest China, aboard a Long March B rocket.
Tianwen-2’s target is a small near-Earth asteroid called 469219 Kamoʻoalewa, which is between 15-39 million km away and is known as a “quasi-satellite” of Earth.
The mission is set to reach the body, which is between 40-100 m wide, in July 2026 where it will first study it up close using a suite of 11 instruments including cameras, spectrometers and radar, before aiming to collect about 100 g of material.
This will be achieved via three possible methods. One is via hovering close to the asteroid, the other is using a robotic arm to collect samples from the body while a third is dubbed “touch and go”, which involves gently landing on the asteroid and using drills at the end of each leg to retrieve material.
The collected samples will then be stored in a module that is released and returned to Earth in November 2027. If successful, it will make China the third nation to retrieve asteroid material behind the US and Japan.
Next steps
The second part of the 10-year mission involves using Earth for a gravitational swing-by to spend six year travelling to another target — 311P/PanSTARRS. The body lies in the main asteroid belt between Mars and Jupiter and at its closest distance is about 140 million km away from Earth.
The 480 m-wide object, which was discovered in 2013, has six dust tails and has characteristics of both asteroids and comets. Tianwen-2 will not land on 311P/PanSTARRS but instead use its instruments to study the “active asteroid” from a distance.
Tianwen-2’s predecessor, Tianwen-1, was China’s first mission to Mars, successfully landing on Utopia Planitia — a largely flat impact basin but scientifically interesting with potential water-ice underneath – following a six-month journey.
China’s third interplanetary mission, Tianwen-3, will aim to retrieve sample from Mars and could launch as soon as 2028. If successful, it would make China the first country to achieve the feat.
Researchers in China have adapted the interlocking structure of mortise-and-tenon joints – as used by woodworkers around the world since ancient times – to the design of nanoscale devices known as memristors. The new devices are far more uniform than previous such structures, and the researchers say they could be ideal for scientific computing applications.
The memory-resistor, or “memristor”, was described theoretically at the beginning of the 1970s, but the first practical version was not built until 2008. Unlike standard resistors, the resistance of a memristor changes depending on the current previously applied to it, hence the “memory” in its name. This means that a desired resistance can be programmed into the device and subsequently stored. Importantly, the remembered value of the resistive state persists even when the power is switched off.
Thanks to numerous technical advances since 2008, memristors can now be integrated onto chips in large numbers. They are also capable of processing large amounts of data in parallel, meaning they could be ideal for emerging “in-memory” computing technologies that require calculations known as large-scale matrix-vector multiplications (MVMs). Many such calculations involve solving partial differential equations (PDEs), which are used to model complex behaviour in fields such as weather forecasting, fluid dynamics and astrophysics, to name but a few.
One remaining hurdle, however, is that it is hard to make memristors with uniform characteristics. The electronic properties of devices containing multiple memristors can therefore vary considerably, which adversely affects the computational accuracy of large-scale arrays.
Inspiration from an ancient technique
Physicists co-led by Shi-Jun Liang and Feng Miao of Nanjing University’s School of Physics say they have now overcome this problem by designing a memristor that uses a mortise-tenon-shaped (MTS) architecture. Humans have been using these strong and stable structures in wooden furniture for thousands of years, with one of the earliest examples dating back to the Hemudu culture in China 7 000 years ago.
Liang, Miao and colleagues created the mortise part of their structure by using plasma etching to create a hole within a nanosized-layer of hexagonal boron nitride (h-BN). They then constructed a tenon in a top electrode made of tantalum (Ta) that precisely matches the mortise. This ensures that this electrode directly contacts the device’s switching layer (which is made from HfO2) only in the designated region. A bottom electrode completes the device.
The new architecture ensures highly uniform switching within the designated mortise-and-tenon region, resulting in a localized path for electronic conduction. “The result is a memristor with exceptional fundamental properties across three key metrics,” Miao tells Physics World. “These are: high endurance (over more than 109 cycles); long-term and stable memory retention (of over 104 s), and a fast switching speed of around 4.2 ns.”
The cycle-to-cycle variation of the low-resistance state (LRS) can also be reduced from 30.3% for a traditional memristor to 2.5% for the MTS architecture and the high-resistance state (HRS) from 62.4 to 27.2%.
To test their device, the researchers built a PDE solver with it. They found that their new MTS memristor could solve the Poisson equation five times faster than a conventional memristor based on HfO2 without h-BN.
The new technique, which is detailed in Science Advances, is a promising strategy for developing high-uniformity memristors, and could pave the way for high-accuracy, energy-efficient scientific computing platforms, Liang claims. “We are now looking to develop large-scale integration of our MTS device and make a prototype system,” he says.
On April 28, Spain experienced one of the most extensive power outages in recent memory. Millions of citizens and businesses were suddenly cut off, revealing how unprepared even developed nations […]
A new contact lens enables humans to see near-infrared light without night vision goggles or other bulky equipment. The lens, which incorporates metallic nanoparticles that “upconvert” normally-invisible wavelengths into visible ones, could have applications for rescue workers and others who would benefit from enhanced vision in conditions with poor visibility.
The infrared (IR) part of the electromagnetic spectrum encompasses light with wavelengths between 700 nm and 1 mm. Human eyes cannot normally detect these wavelengths because opsins, the light-sensitive protein molecules that allow us to see, do not have the required thermodynamic properties. This means we see only a small fraction of the electromagnetic spectrum, typically between 400‒700 nm.
While devices such as night vision goggles and infrared-visible converters can extend this range, they require external power sources. They also cannot distinguish between different wavelengths of IR light.
Photoreceptor-binding nanoparticles
In a previous work, researchers led by neuroscientist Tian Xue of the University of Science and Technology of China (USTC) injected photoreceptor-binding nanoparticles into the retinas of mice. While this technique was effective, it is too invasive and risky for human volunteers. In the new study, therefore, Xue and colleagues integrated the nanoparticles into biocompatible polymeric materials similar to those used in standard soft contact lenses.
The nanoparticles in the lenses are made from Au/NaGdF4: Yb3+, Er3+ and have a diameter of approximately 45 nm each. They work by capturing photons with lower energies (longer wavelengths) and re-emitting them as photons with higher energies (shorter wavelengths). This process is known as upconversion and the emitted light is said to be anti-Stokes shifted.
When the researchers tested the new upconverting contact lenses (UCLs) on mice, the rodents’ behaviour suggested they could sense IR wavelengths. For example, when given a choice between a dark box and an IR-illuminated one, the lens-wearing mice scurried into the dark box. In contrast, a control group of mice not wearing lenses showed no preference for one box over the other. The pupils of the lens-wearing mice also constricted when exposed to IR light, and brain imaging revealed that processing centres in their visual cortex were activated.
Flickering seen even with eyes closed
The team then moved on to human volunteers. “In humans, the near-infrared UCLs enabled participants to accurately detect flashing Morse code-like signals and perceive the incoming direction of near-infrared (NIR) light,” Xue says, referring to light at wavelengths between 800‒1600 nm. Counterintuitively, the flashing images appeared even clearer when the volunteers closed their eyes – probably because IR light is better than visible light at penetrating biological tissue such as eyelids. Importantly, Xue notes that wearing the lenses did not affect participants’ normal vision.
The team also developed a wearable system with built-in flat UCLs. This system allowed volunteers to distinguish between patterns such as horizontal and vertical lines; S and O shapes; and triangles and squares.
But Xue and colleagues did not stop there. By replacing the upconverting nanoparticles with trichromatic orthogonal ones, they succeeded in converting NIR light into three different spectral bands. For example, they converted infrared wavelengths of 808, 980 nm and 1532 nm into 540, 450, and 650 nm respectively – wavelengths that humans perceive as green, blue and red.
“As well as allowing wearers to garner more detail within the infrared spectrum, this technology could also help colour-blind individuals see wavelengths they would otherwise be unable to detect by appropriately adjusting the absorption spectrum,” Xue tells Physics World.
According to the USTC researchers, who report their work in Cell, the devices could have several other applications. Apart from providing humans with night vision and offering an adaptation for colour blindness, the lenses could also give wearers better vision in foggy or dusty conditions.
At present, the devices only work with relatively bright IR emissions (the study used LEDs). However, the researchers hope to increase the photosensitivity of the nanoparticles so that lower levels of light can trigger the upconversion process.
Astronstone, one of China’s newest commercial launch startups, has raised early-stage funding for a stainless steel, reusable launch vehicle modeled on SpaceX’s Starship system.
Powerful flares on highly-magnetic neutron stars called magnetars could produce up to 10% of the universe’s gold, silver and platinum, according to a new study. What is more, astronomers may have already observed this cosmic alchemy in action.
Gold, silver, platinum and a host of other rare heavy nuclei are known as rapid-process (r-process) elements. This is because astronomers believe that these elements are produced by the rapid capture of neutrons by lighter nuclei. Neutrons can only exist outside of an atomic nucleus for about 15 min before decaying (except in the most extreme environments). This means that the r-process must be fast and take place in environments rich in free neutrons.
In August 2017, an explosion resulting from the merger of two neutron stars was witnessed by telescopes operating across the electromagnetic spectrum and by gravitational-wave detectors. Dubbed a kilonova, the explosion produced approximately 16,000 Earth-masses worth of r-process elements, including about ten Earth masses of gold and platinum.
While the observations seem to answer the question of where precious metals came from, there remains a suspicion that neutron-star mergers cannot explain the entire abundance of r-process elements in the universe.
Giant flares
Now researchers led by Anirudh Patel, who is a PhD student at New York’s Columbia University, have created a model that describes how flares on the surface of magnetars can create r-process elements.
Patel tells Physics World that “The rate of giant flares is significantly greater than mergers.” However, given that one merger “produces roughly 10,000 times more r-process mass than a single magnetar flare”, neutron-star mergers are still the dominant factory of rare heavy elements.
A magnetar is an extreme type of neutron star with a magnetic field strength of up to a thousand trillion gauss. This makes magnetars the most magnetic objects in the universe. Indeed, if a magnetar were as close to Earth as the Moon, its magnetic field would wipe your credit card.
Astrophysicists believe that when a magnetar’s powerful magnetic fields are pulled taut, the magnetic tension will inevitably snap. This would result in a flare, which is an energetic ejection of neutron-rich material from the magnetar’s surface.
Mysterious mechanism
However, the physics isn’t entirely understood, according to Jakub Cehula of Charles University in the Czech Republic, who is a member of Patel’s team. “While the source of energy for a magnetar’s giant flares is generally agreed to be the magnetic field, the exact mechanism by which this energy is released is not fully understood,” he explains.
One possible mechanism is magnetic reconnection, which creates flares on the Sun. Flares could also be produced by energy released during starquakes following a build-up of magnetic stress. However, neither satisfactorily explains the giant flares, of which only nine have thus far been detected.
In 2024 Cehula led research that attempted to explain the flares by combining starquakes with magnetic reconnection. “We assumed that giant flares are powered by a sudden and total dissipation of the magnetic field right above a magnetar’s surface,” says Cehula.
This sudden release of energy drives a shockwave into the magnetar’s neutron-rich crust, blasting a portion of it into space at velocities greater than a tenth of the speed of light, where in theory heavy elements are formed via the r-process.
Gamma-ray burst
Remarkably, astronomers may have already witnessed this in 2004, when a giant magnetar flare was spotted as a half-second gamma-ray burst that released more energy than the Sun does in a million years. What happened next remained unexplained until now. Ten minutes after the initial burst, the European Space Agency’s INTEGRAL satellite detected a second, weaker signal that was not understood.
Now, Patel and colleagues have shown that the r-process in this flare created unstable isotopes that quickly decayed into stable heavy elements – creating the gamma-ray signal.
Patel calculates that the 2004 flare resulted in the creation of two million billion billion kilograms of r-process elements, equivalent to about the mass of Mars.
Extrapolating, Patel calculates that giant flares on magnetars contribute between 1–10% of all the r-process elements in the universe.
Lots of magnetars
“This estimate accounts for the fact that these giant flares are rare,” he says, “But it’s also important to note that magnetars have lifetimes of 1000 to 10,000 years, so while there may only be a couple of dozen magnetars known to us today, there have been many more magnetars that have lived and died over the course of the 13 billion-year history of our galaxy.”
Magnetars would have been produced early in the universe by the supernovae of massive stars, whereas it can take a billion years or longer for two neutron stars to merge. Hence, magnetars would have been a more dominant source of r-process elements in the early universe. However, they may not have been the only source.
“If I had to bet, I would say there are other environments in which r-process elements can be produced, for example in certain rare types of core-collapse supernovae,” says Patel.
Either way, it means that some of the gold and silver in your jewellery was forged in the violence of immense magnetic fields snapping on a dead star.
Learn how ice crystallizes on the surface of Europa, a moon of Jupiter, may hold promise for signs of extraterrestrial life within its subsurface ocean.
Quantum science is enjoying a renaissance as nascent quantum computers emerge from the lab and quantum sensors are being used for practical applications.
As the technologies we use become more quantum in nature, it follows that everyone should have a basic understanding of quantum physics. To explore how quantum physics can be taught to the masses, I am joined by Arjan Dhawan, Aleks Kissinger and Bob Coecke – who are all based in the UK.
Coecke is chief scientist at Quantinuum – which develops quantum computing hardware and software. Kissinger is associate professor of quantum computing at the University of Oxford; and Dhawan is studying mathematics at the University of Durham.
Kissinger and Coecke have developed a way of teaching quantum physics using diagrams. In 2023, Oxford and Quantinuum joined forces to use the method in a pilot summer programme for 15 to 17 year-olds. Dhawan was one of their students.
In this week's episode of Space Minds Ashley Johnson, President and CFO of Planet, explains the company's ambitious goal to make global change visible, accessible and actionable.
Adaptive radiotherapy, an advanced cancer treatment in which each fraction is tailored to the patient’s daily anatomy, offers the potential to maximize target conformality and minimize dose to surrounding healthy tissue. Based on daily scans – such as MR images recorded by an MR-Linac, for example – treatment plans are adjusted each day to account for anatomical changes in the tumour and surrounding healthy tissue.
Creating a new plan for every treatment fraction, however, increase the potential for errors, making fast and effective quality assurance (QA) procedures more important than ever. To meet this need, the physics team at Hospital Almater in Mexicali, Mexico, is using Elekta ONE | QA, powered by ThinkQA Secondary Dose Check* (ThinkQA SDC) software to ensure that each adaptive plan is safe and accurate before it is delivered to the patient.
Radiotherapy requires a series of QA checks prior to treatment delivery, starting with patient-specific QA, where the dose calculated by the treatment planning system is delivered to a phantom. This procedure ensures that the delivered dose distribution matches the prescribed plan. Alongside, secondary dose checks can be performed, in which an independent algorithm verifies that the calculated dose distribution corresponds with that delivered to the actual patient anatomy.
“The secondary dose check is an independent dose calculation that uses a different algorithm to the one in the treatment planning system,” explains Alexis Cabrera Santiago, a medical physicist at Hospital Almater. “ThinkQA SDC software calculates the dose based on the patient anatomy, which is actually more realistic than using a rigid phantom, so we can compare both results and catch any differences before treatment.”
Pre-treatment verification ThinkQA SDC’s unique dose calculation method has been specifically designed for Elekta Unity. (Courtesy: Elekta)
For adaptive radiotherapy in particular, this second check is invaluable. Performing phantom-based QA following each daily imaging session is often impractical. Instead, in many cases, it’s possible to use ThinkQA SDC instead.
“Secondary dose calculation is necessary in adaptive treatments, for example using the MR-Linac, because you are changing the treatment plan for each session,” says José Alejandro Rojas‑López, who commissioned and validated ThinkQA SDC at Hospital Almater. “You are not able to shift the patient to realise patient-specific QA, so this secondary dose check is needed to analyse each treatment session.”
ThinkQA SDC’s ability to achieve patient-specific QA without shifting the patient is extremely valuable, allowing time savings while upholding the highest level of QA safety. “The AAPM TG 219 report recognises secondary dose verification as a validated alternative to patient-specific QA, especially when there is no time for traditional phantom checks in adaptive fractions,” adds Cabrera Santiago.
The optimal choice
At Hospital Almater, all external-beam radiation treatments are performed using an Elekta Unity MR-Linac (with brachytherapy employed for gynaecological cancers). This enables the hospital to offer adaptive radiotherapy for all cases, including head-and-neck, breast, prostate, rectal and lung cancers.
To ensure efficient workflow and high-quality treatments, the team turned to the ThinkQA SDC software. ThinkQA SDC received FDA 510(k) clearance in early 2024 for use with both the Unity MR-Linac and conventional Elekta linacs.
Rojas‑López (who now works at Hospital Angeles Puebla) says that the team chose ThinkQA SDC because of its user-friendly interface, ease of integration into the clinical workflow and common integrated QA platform for both CT and MR-Linac systems. The software also offers the ability to perform 3D evaluation of the entire planning treatment volume (PTV) and the organs-at-risk, making the gamma evaluation more robust.
Physics team Alexis Cabrera Santiago and José Alejandro Rojas‑López. (Courtesy: José Alejandro Rojas‑López/Hospital Almater)
Commissioning of ThinkQA SDC was fast and straightforward, Rojas‑López notes, requiring minimal data input into the software. For absolute dose calibration, the only data needed are the cryostat dose attenuation response, the output dose geometry and the CT calibration.
“This makes a difference compared with other commercial solutions where you have to introduce more information, such as MLC [multileaf collimator] leakage and MLC dosimetric leaf gap, for example,” he explains. “If you have to introduce more data for commissioning, this delays the clinical introduction of the software.”
Cabrera Santiago is now using ThinkQA SDC to provide secondary dose calculations for all radiotherapy treatments at Hospital Almater. The team has established a protocol with a 3%/2 mm gamma criterion, a tolerance limit of 95% and an action limit of 90%. He emphasizes that the software has proved robust and flexible, and provides confidence in the delivered treatment.
“ThinkQA SDC lets us work with more confidence, reduces risk and saves time without losing control over the patient’s safety,” he says. “It checks that the plan is correct, catches issues before treatment and helps us find any problems like set-up errors, contouring mistakes and planning issues.”
The software integrates smoothly into the Elekta ONE adaptive workflow, providing reliable results without slowing down the clinical workflow. “In our institution, we set up ThinkQA SDC so that it automatically receives the new plan, runs the check, compares it with the original plan and creates a report – all in around two minutes,” says Cabrera Santiago. “This saves us a lot of time and removes the need to do everything manually.”
A case in point
As an example of ThinkQA SDC’s power to ease the treatment workflow, Rojas‑López describes a paediatric brain tumour case at Hospital Almater. The young patient needed sedation during their treatment, requiring the physics team to optimize the treatment time for the entire adaptive radiotherapy workflow. “ThinkQA SDC served to analyse, in a fast mode, the treatment plan QA for each session. The measurements were reliable, enabling us to deliver all of the treatment sessions without any delay,” he explains.
Indeed, the ability to use secondary dose checks for each treatment fraction provides time advantages for the entire clinical workflow over phantom-based pre-treatment QA. “Time in the bunker is very expensive,” Rojas‑López points out. “If you reduce the time required for QA, you can use the bunker for patient treatments instead and treat more patients during the clinical time. Secondary dose check can optimize the workflow in the entire department.”
The team at Hospital Almater concur that ThinkQA SDC provides a reliable tool to evaluate radiation treatments, including the first fraction and all of the adaptive sessions, says Rojas‑López. “You can use it for all anatomical sites, with reliable and confident results,” he notes. “And you can reduce the need for measurements using another patient-specific QA tool.”
“I think that any centre doing adaptive radiotherapy should seriously consider using a tool like ThinkQA SDC,” adds Cabrera Santiago.
*ThinkQA is manufactured by DOSIsoft S.A. and distributed by Elekta.
Chinese rocket maker Sepoch has carried out a first vertical liftoff and splashdown landing ahead of a potential orbital launch attempt later this year.
Off-world agriculture has long seemed experimental, but that could soon change thanks to a collaboration between design firm Heatherwick Studio and the space architecture nonprofit Aurelia Institute.
Anna’s hummingbirds have evolved to have longer, larger beaks to access backyard feeders in urban areas. It could be a step toward becoming a “commensal” species that lives alongside humans, like pigeons.
The first high-resolution images of Bolivia’s Uturuncu volcano have yielded unprecedented insights into whether this volcanic “zombie” is likely to erupt in the near future. The images were taken using a technique that combines seismology, rock physics and petrological analyses, and the scientists who developed it say it could apply to other volcanoes, too.
Volcanic eruptions occur when bubbles of gases such as SO2 and CO2 rise to the Earth’s surface through dikes and sills in the planet’s crust, bringing hot, molten rock known as magma with them. To evaluate the chances of this happening, researchers need to understand how much gas and melted rock have accumulated in the volcano’s shallow upper crust, or crater. This is not easy, however, as the structures that convey gas and magma to the surface are complex and mapping them is challenging with current technologies.
A zombie volcano
In the new work, a team led by Mike Kendall of the University of Oxford, UK and Haijiang Zhang from the University of Science and Technology of China (USTC) employed a combination of seismological and petrophysical analyses to create such a map for Uturuncu. Located in the Central Andes, this volcano formed in the Pleistocene era (around 2.58 million to 11,700 years ago) as the oceanic Nazca plate was forced beneath the South American continental plate. It is made up of around 50 km3 of homogeneous, porphyritic dacite lava flows that are between 62% and 67% silicon dioxide (SiO2) by weight, and it sits atop the Altiplano–Puna magma body, which is the world’s largest body of partially-melted silicic rock.
Although Uturuncu has not erupted for nearly 250,000 years, it is not extinct. It regularly emits plumes of gas, and earthquakes are a frequent occurrence in the shallow crust beneath and around it. Previous geodetic studies also detected a 150-km-wide deformed region of rock centred around 3 km south-west of its summit. These signs of activity, coupled with Uturuncu’s lack of a geologically recent eruption, have led some scientists to describe it as a “zombie”.
Movement of liquid and gas explains Uturuncu’s unrest
To tease out the reasons for Uturuncu’s semi-alive behaviour, the team turned to seismic tomography – a technique Kendall compares to medical imaging of a human body. The idea is to detect the seismic waves produced by earthquakes travelling through the Earth’s crust, analyse their arrival times, and use this information to create three-dimensional images of what lies beneath the surface of the structure being studied.
Writing in PNAS, Kendall and colleagues explain that they used seismic tomography to analyse signals from more than 1700 earthquakes in the region around Uturuncu. They performed this analysis in two ways. First, they assumed that seismic waves travel through the crust at the same speed regardless of their direction of propagation. This isotropic form of tomography gave them a first image of the region’s structure. In their second analysis, they took the directional dependence of the seismic waves’ speed into account. This anisotropic tomography gave them complementary information about the structure.
The researchers then combined their tomographic measurements with previous geophysical imaging results to construct rock physics models. These models contain information about the paths that hot migrating fluids and gases take as they migrate to the surface. In Uturuncu’s case, the models showed fluids and gases accumulating in shallow magma reservoirs directly below the volcano’s crater and down to a depth of around 5 km. This movement of liquid and gas explains Uturuncu’s unrest, the team say, but the good news is that it has a low probability of producing eruptions any time soon.
According to Kendall, the team’s methods should be applicable to more than 1400 other potentially active volcanoes around the world. “It could also be applied to identifying potential geothermal energy sites and for critical metal recovery in volcanic fluids,” he tells Physics World.
NASA has decided to switch to a backup propellant line on its Psyche asteroid mission to allow the spacecraft to resume use of its electric propulsion system.