Todd McNutt is a radiation oncology physicist at Johns Hopkins University in the US and the co-founder of Oncospace, where he led the development of an artificial intelligence (AI)-powered tool that simultaneously accelerates radiation planning and elevates plan quality and consistency. The software, now rebranded as Plan AI and available from US manufacturer Sun Nuclear, draws upon data from thousands of previous radiotherapy treatments to predict the lowest possible dose to healthy tissues for each new patient. Treatment planners then use this information to define goals that streamline and automate the creation of a best achievable plan.

Physics World’s Tami Freeman spoke with McNutt about the evolution of Oncospace and the benefits that Plan AI brings to radiotherapy patients and cancer treatment centres.

Can you describe how the Oncospace project began?

Back in 2007, several groups were discussing how we could better use clinical data for discovery and knowledge generation. I had several meetings with folks at Johns Hopkins, including Alex Szalay who helped develop the Sloan Digital Sky Survey. He built a large database of galaxies and stars and it became a huge research platform for both amateur and professional astronomers.

From that discussion, and other initiatives, we looked at moving towards structured data collection for patients in the clinical environment. By marrying these data with radiation treatment plans we could study how dose distributions across the anatomy affect patient outcomes. And we took that opportunity to build a database for radiotherapy.

What inspired the transition from academic research to founding the company Oncospace Inc in 2019?

After populating the database with data from many patients, we could examine which anatomic features impact our ability to generate a plan that minimizes radiation dose to normal tissues while treating target volumes as best as possible. We came up with a feature set that characterized the relationships between normal anatomy and targets, as well as target complexity.

This early work allowed us to predict expected doses from these shape-relationship features, and it worked well. At that point, we knew we could tap into this database and generate a prediction that could help create treatment plans for new patients. We thought of this as personalized medicine: for the first time, we could see the level of treatment plan quality that we could achieve for a specific patient.

I thought that this was useful commercially and that we should get it out to other clinics. Praveen Sinha, who I’d known from my previous work at Philips and now leads Sun Nuclear’s software business line, asked if I wanted to create a startup. The timing was right for both of us and I had a team here ready to go, so we went ahead and did it. With his knowledge of startups and my knowledge of what we wanted to achieve, we had perfect timing and a perfect group to work with.

Plan AI enables both predictive planning and peer review, how do these functions work?

The idea behind predictive planning is that, for a given patient, I can predict the expected dose that I should be able to achieve for them.

Treatment planning involves specifying dosimetric objectives to the planning system and asking it to optimize radiation delivery to meet these. But nobody really knows what the right objectives even are – it is just a trial-and-error process. Plan AI’s prediction provides a rational set of objectives for plan optimization, allowing the planning system’s algorithm to move towards a good solution and making treatment planning an easier problem to solve.

Peer review involves a peer physician looking at every treatment plan to evaluate it for quality and safety. But again, people don’t really know the level of quality you can generate, it depends on the patient’s anatomy. Providing a predicted dose with clinical dose goals enables a rapid review to see whether it is a high-quality plan or not.

In the past we looked at simple things like whether a contour is missing slices or contains discontinuities and Plan AI checks for this, but you can do far more with AI. For example, you could look at all the contoured rectums in the system and predict if your contour goes too far into the sigmoid colon, then it may be mis-contoured. We have research software that can flag such potential anomalies so they don’t get overlooked.

The Plan AI models are developed using Oncospace’s database of previous treatments; can you describe this data lake?

When we first started, we developed a large SQL database containing all the shape-relationship features and dosimetry features. The SQL language is ideal for being able to query and sift through the data, but when the company was formed, we recognized that there was some age to that technology.

So for the Plan AI data lake, we extracted all the different shape-relationship and shape-complexity features and put them into a Parquet database in the cloud. This made the data lake much more amenable to applying machine learning algorithms to it. The SQL data lake at Johns Hopkins is maintained separately and primarily used to investigate toxicity predictions and spatial dose patterns. But for Plan AI, the models are fixed and streamlined for the specific task of dose prediction.

What does the model training process entail?

One of the first tasks was to curate the data, using the AAPM’s standardized structure-naming model. Our data scientist Julie Shade wrote some tools for automatic name mapping and target identification; that helped us process much larger amounts of data for the model.

Once we had all the shape-relationship and shape-complexity features and all the doses, we trained the models by anatomical region. We have FDA-approved models for the male and female pelvis, thorax, abdomen and head-and-neck. For each of these, we predict the doses for every organ-at-risk. Then we used a five-fold validation model to make sure that the predictions were good on an internal data set.

We also performed external validation at institutions including Johns Hopkins and Montefiore hospitals. We created predicted plans from recent treatment plans that had been evaluated by physicians. For almost all cases, both plan quality and plan efficiency were improved with Plan AI.

One aspect of this training is that whenever we drive optimization via predictive planning we want to push towards the best achievable dose. Regular machine learning predicts an expected, or average, dose across all patients. But you never want to drive a treatment plan towards the average dose, because then every plan you generate will be happy being average. Our model predicts both the average and the best achievable dose, and drives plan optimization towards the best achievable.

When implementing new technology in the clinic, it’s important to fit into the existing treatment workflow. How clinic-ready are these AI tools?

Radiation therapy is protocol-driven: we know what technique we’re going to use to treat and what our clinical dose goals are for different structures. What we don’t know is the patient-specific part of that. So for each anatomical region, we built models out of a wide range of treatment protocols, with many different types of patients, to ensure that the same prediction model works for any protocol. This means a user can use any protocol for treatment and the predictions will work, they don’t have to retrain anything. It’s ready to go out of the box, there’s a library of protocols to start with, and you can change protocols as you need for your own clinic.

The other part of being clinic-ready is aligning with the way that planning is currently performed, which is using dose–volume histograms. Treatment plans are optimized by manipulating these dose objectives, and that’s exactly what we predict. So users aren’t changing the whole paradigm of how planners operate. They still use their treatment planning system (TPS) – we just put the objectives in there. Basically, a TPS script sends the patient’s CT and contours to the cloud, where Plan AI makes the predictions. The TPS then pulls back in the objectives built from the models, based on this specific patient’s anatomy. The TPS runs the optimization and, as a last step, can send the plan back to Plan AI to check that it fits within the best achievable predictions.

Did you encounter any challenges bringing AI into a clinical setting?

Interestingly, the challenges aren’t technical, they are more human related. One of the more systemic challenges is data security when using medical data for training. A nice thing about our system is that the features we generate from treatment plans are just mathematical shape-relationship features and don’t involve a lot of identifiable information.

AI has been used in radiation therapy for image contouring and auto-segmentation, and early efforts were not so good. So, there’s always a good, healthy scepticism. But once you show people that it works and works well, this can be overcome. I have seen some people worried about job security and AI taking over. We are medical professionals designing a treatment plan to care for a patient and there’s a lot of pride and art in that – if you automate that, it takes away some of this pride and art.

I tell people that if we automate the easier things, then they can spend their quality time on the more difficult and challenging cases, because that’s where their talent might be needed more.

Do you have any advice for clinics looking to adopt AI-driven planning?

Introduce it as an assistant, not as a solution. You want people that already know what they’re doing to be able to use their knowledge more efficiently. We want to make their jobs easier and show them that it also improves quality.

With dosimetrists, for example, they create a plan and work hard getting the dose down – and then the physician looks at it and suggests that they can do better. Predictive planning gives them confidence that they are right and takes the uncertainty out of the physician review process. And once you’ve gained that level of confidence, you can start using it for adaptive planning or other technologies.

Where do you see predictive modelling and AI in oncology in five years from now?

Right now, there’s been a lot of data collected, but we want that data to advance and learn. Having multiple centres adding to this pool of knowledge and being able to continually update those models from new, broader data sets could be of huge value.

In terms of patient outcomes, we’ve done a lot of the work looking at how the spatial pattern of dose impacts toxicity and outcomes. This is part of the research being performed at Johns Hopkins and still in discovery mode. But down the road, some of these predictions of normal tissue outcomes could be fed into the planning process to help reduce toxicity at the patient level.

Finally, what’s been the most rewarding part of this journey for you?

During my prior experience building treatment planning systems, the biggest problem was always that nobody knew what the objective was. Nobody knew how to tell the system: “this is the dose I expect to receive, now optimize to get it for me”, because you didn’t know what you could do. For any given patient, you could ask for too much or too little. Now, for the first time, I argue that we actually know what our objective is in our treatment planning.

This levels the playing field between different environments, different countries, or even different dosimetrists with different levels of experience. The Plan AI tool brings all this to a consistent state and enables high quality, efficient planning everywhere. We can provide this predictive planning tool to clinics around the world. Now we just have to get everybody using it.

• You can listen to the full interview with Todd McNutt in the Physics World Weekly podcast.

 

The post Todd McNutt: how an AI software solution enables creation of the best possible radiation treatment plans appeared first on Physics World.

In his opening remarks to the 4th International Symposium on the History of Particle Physics, Chris Llewellyn Smith – who was a director-general of CERN in the 1990s – suggested participants should speak about “what’s not written in the journals”, including “mistakes, dead-ends and problems with getting funding”. Doing so, he said, would “provide insight into the way science really progresses”.

The symposium was not your usual science conference. Held last November at CERN, it took place inside the lab’s 400-seat main auditorium, which has been the venue for many historic announcements, including the discovery of the Higgs boson. Its brown-beige walls are covered with lively designs by the Finnish artist Ilona Rista, suggesting to me the aftermath of a collision of high-energy bar codes.

The 1980s and 1990s saw the construction and operation of various important accelerators and detectors.

The focus of the meeting was the development of particle physics in the 1980s and 1990s – a period that saw the construction and operation of various important accelerators and detectors. At CERN, these included the UA1 and UA2 experiments at the Super Proton Synchrotron, where the W and Z bosons were discovered. Later, there was the Large Electron-Positron Collider (LEP), which came online in 1989, and the Large Hadron Collider (LHC), approved five years later.

Delegates also heard about the opening of various accelerators in the US during those two decades, including two at the Stanford Linear Accelerator Center – the Positron-Electron Project in 1980 and the Stanford Linear Collider in 1989. Most famous of all was the start-up of the Tevatron at Fermilab in 1983. Over at Dubna in the former Soviet Union, meanwhile, scientists built the Nuclotron, a superconducting synchrotron, which opened in 1992.

Conference speakers covered unfinished machines of the era as well. The US cancelled two proton–proton facilities – ISABELLE in 1983 and the Superconducting Super Collider (SSC) a decade later. The Soviet Union, meanwhile, abandoned the multi-TeV proton–proton collider UNK a few years later, though news has recently emerged that Russia might revive the project.

Several speakers recounted the discovery of the W and Z particles at CERN in 1983 and the discovery of the top quark at Fermilab in 1995. Others addressed the strange fact that fewer neutrinos from the Sun had been detected than theory suggested. The “solar-neutrino problem”, as it was known, was finally resolved by Takaaki Kajita’s discovery of neutrino oscillation in 1998, for which he shared the 2015 Nobel Prize for Physics with Art McDonald.

The conference also addressed unsuccessful searches for proton decay, axions, magnetic monopoles, the Higgs boson, supersymmetry particles and other targets. Other speakers described projects with highly positive outcomes, such as the advent of particle cosmology, or what some have jokingly dubbed “the heavenly lab”. The development of string theory, grand unified theories and perturbative quantum chromodynamics was tackled too.

In an exchange in the question-and-answer session after one talk, the Greek physicist Kostas Gavroglu referred to many of such quests as “failures”. That remark prompted the Australian-born US theoretical physicist Helen Quinn to say she preferred the term “falling forward”; such failures, she said, were instances of “I tried this, and it didn’t work so I tried that”.

In relating his work on detecting gravitational waves, the US Nobel-prize-winning physicist Barry Barish said he felt his charge was not to celebrate the importance of his discoveries nor the ingenuity of the route he took. Instead, Barish explained, his job was to answer the much more informal question: “What made me do what?”.

His point was illustrated by the US theorist Alan Guth, who described the very human and serendipitous path he took to working on cosmic inflation – the super-fast expansion of the universe just after the Big Bang. When he started, Guth said, “all the ingredients were already invented”. But the startling idea of inflation hinged on accidental meetings, chance conversations, unexpected visits, a restricted word count for Physical Review Letters, competitions, insecurities and “spectacular realizations” coalescing.

Wider world

Another theme that arose in the conference was that science does not unfold inside its own bubble but can have extensive and immediate impacts on the world around it. Two speakers, for instance, recounted the invention of the World Wide Web at CERN in the late 1980s. It’s fair to say that no other discovery by a single individual – Tim Berners-Lee – has so radically and quickly transformed the world.

The growing role of international politics in promoting and protecting projects was mentioned too, with various speakers explaining how high-level political negotiations enabled physicists to work at institutions and experiments in other nations. The Polish physicist Agnieszka Zalewska, for example, described her country’s path to membership in CERN, while Russian-born US physicist Vladimir Shiltsev spoke about the “diaspora” of Russian particle physicists after the fall of the Soviet Union in 1991.

As a result of the Superconducting Super Colllider’s controversial closure, the centre of gravity of high-energy physics shifted to Europe.

Sometimes politics created destructive interference. The US physicist, historian and author Michael Riordan described how the US’s determination to “go it alone” to outcompete Europe in high-energy physics was a major factor in bringing about the opposite: the termination of the SSC in 1993. As a result of that project’s controversial closure, the centre of gravity of high-energy physics shifted to Europe.

Indeed, contemporary politics occasionally hit the conference itself in incongruous and ironic ways. Two US physicists, for example, were denied permission to attend because budgets had been cut and travel restrictions increased. In the end, one took personal time off and paid his own way, leaving his affiliation off the programme.

Before the conference, some people complained that conference organizers hadn’t paid enough attention to physicists who’d worked in the Soviet Union but were from occupied republics. Several speakers addressed this shortcoming by mentioning people like Gersh Budker (1918–1977). A Ukrainian-born physicist who worked and died in the Soviet Union, Budker was nominated for a Nobel Prize (1957) and even has had a street named after him at CERN. Unmentioned, though, was that Budker was Jewish and that his father was killed by Ukrainian nationalists in a pogrom.

On the final day of the conference, which just happened to be World Science Day for Peace and Development, CERN mounted a public screening of the 2025 documentary film The Peace Particle. Directed by Alex Kiehl, much of it was about CERN’s internationalism, with a poster for the film describing the lab as “Mankind’s biggest experiment…science for peace in a divided world”.

But in the Q&A afterwards some audience members criticized CERN for allegedly whitewashing Russia for its invasion of the Ukraine and Israel for genocide. Those onstage defended CERN on the grounds of its desire to promote internationalism.

The critical point

The keynote speaker of the conference was John Krige, a science historian from Georgia Tech who has worked on a three-volume history of CERN. Those who launched the lab, Krige reminded the audience, had radical “scientific, political and cultural aspirations” for the institution. Their dream was that CERN wouldn’t just revive European science and promote regional collaborative effects after the Second World War, but also potentially improve the global world order too.

Krige went on to quote one CERN founder, who’d said that international science facilities such as CERN would be “one of the best ways of saving Western civilization”. Recent events, however, have shown just how fragile those ambitions are. Alluding to CERN’s Future Circular Collider and other possible projects, Llewellyn Smith ended his closing remarks with a warning.

“The perennial hope that the next big high-energy project will be genuinely global,” he said, “seems to be receding over the horizon due to the polarization of world politics”.

The post The future of particle physics: what can the past teach us? appeared first on Physics World.

Attempts to understand quantum phase transitions in open systems usually rely on real‑time Lindbladian evolution, which tracks how a quantum state changes as it relaxes toward a steady state. This approach is powerful for studying decoherence, dissipation and long‑time behaviour, but it often fails to reveal the deeper structure of the system including the phase transitions, critical points and hidden quantum order that define its underlying physics.

In this work, the researchers introduce a new framework called imaginary‑time Lindbladian evolution, which allows them to define and classify quantum phases in open systems using the spectrum of an imaginary‑Liouville superoperator. This approach works not only for pure ground states but also for finite‑temperature Gibbs states of stabilizer Hamiltonians, showing its relevance for realistic, mixed‑state conditions.

A key diagnostic in their method is the imaginary‑Liouville gap, the spectral gap between the lowest and next‑lowest decay modes. When this gap closes, the system undergoes a phase transition, a change that is accompanied by diverging correlation lengths and nonanalytic shifts in physical observables. The closing of this gap also coincides with the divergence of the Markov length, a recently proposed indicator of criticality in open quantum systems.

To demonstrate the power of their framework, the researchers map out phase diagrams for systems with

$Z_2^{\sigma }\times Z_2^{\tau }$

symmetry, capturing both spontaneous symmetry breaking and average symmetry‑protected topological phases. Their method reveals universal critical behaviour that real‑time Lindbladian steady states fail to detect, highlighting why imaginary‑time evolution fills a missing piece in the theory of open‑system phases.

Importantly, the authors emphasise that real‑time Lindbladians remain essential for modelling dissipation in practical settings. Their new framework complements this conventional approach, offering a systematic way to study phase transitions in open systems. They also outline how phase diagrams can be constructed using both bottom‑up (state‑based) and top‑down (Hamiltonian‑based) strategies, illustrating the method with a dissipative transverse‑field Ising model.

Overall, this work provides a unified and versatile way to understand quantum phases in open systems, revealing critical behaviour and topological structure that were previously inaccessible. It opens new directions for studying mixed‑state quantum matter and advances the theoretical foundations needed for future quantum technologies.

Do you want to learn more about this topic?

Focus on Quantum Entanglement: State of the Art and Open Questions guest edited by Anna Sanpera and Carlo Marconi (2025-2026)

The post A breakthrough in modelling open quantum matter appeared first on Physics World.

In the macroscopic world, we see irreversible processes everywhere, heat flowing from hot to cold, gases mixing, systems decaying. Yet at the microscopic level, quantum mechanics is perfectly reversible, with its equations running equally well forwards and backwards in time. How then, does irreversibility emerge from fundamentally reversible dynamics?

A common explanation is coarse-graining, which simplifies a complex system by ignoring microscopic details and focusing only on large-scale behaviour. To make the micro–macro divide precise, however, one must first define what “macroscopic” means. Here it is given a quantitative inferential meaning: a state is macroscopic if it is perfectly inferable from the perspective of a specified measurement and prior. Central to this framework is a coarse-graining map built from the measurement and its optimal Bayesian recovery via the Petz map; macroscopic states are precisely its fixed points, turning macroscopicity into a sharp condition of perfect inferability. This construction is grounded in Bayesian retrodiction, which infers what a system likely was before it was measured, together with an observational deficit that quantifies how much information is lost in forming a macroscopic description.

States that are macroscopically inferable can be characterised in several equivalent ways, all tied to to a new measure of disorder called macroscopic entropy, which captures how irreversible, or “uninferable”, a macroscopic process appears from the observer’s perspective. This perspective is formalised through inferential reference frames, built from the combination of a prior and a measurement, which determine what an observer can and cannot recover about the underlying quantum state.

The researchers also develop a resource theory of microscopicity, treating macroscopic states as free and identifying the operations that cannot generate microscopic detail. This unifies and extends existing resource theories of coherence, athermality, and asymmetry. They further introduce observational discord, a new way to understand quantum correlations when observational power is limited, and provide conditions for when this discord vanishes.

Altogether, this work reframes macroscopic irreversibility as an information-theoretic phenomenon, grounded not in a fundamental dynamical asymmetry but in an inferential asymmetry arising from the observer’s limited perspective. It offers a unified way to understand coarse-graining, entropy, and the emergence of classical behaviour from quantum mechanics. It deepens our understanding of time’s direction and has implications for quantum computing, thermodynamics, and the study of quantum correlations in realistic, constrained settings.

Do you want to learn more about this topic?

Focus on Quantum Entanglement: State of the Art and Open Questions guest edited by Anna Sanpera and Carlo Marconi (2025-2026)

The post How reversibility becomes irreversible appeared first on Physics World.

Researchers at Los Alamos National Laboratory in New Mexico, US have used visible light to both image and manipulate the domains of a chiral antiferromagnet (AFM). By “painting” complex patterns onto samples of cobalt niobium sulfite (Co_1/3NbS₂), they demonstrated that it is possible to control AFM domain formation and dynamics, boosting prospects for data storage devices based on antiferromagnetic materials rather than the ferromagnetic ones commonly used today.

In antiferromagnetic materials, the spins of neighbouring atoms in the material’s lattice are opposed to each other (they are antiparallel). For this reason, they do not exhibit a net magnetization in the absence of a magnetic field. This characteristic makes them largely immune to disturbances from external magnetic fields, but it also makes them all but invisible to simple electrical and optical probes, and extremely difficult to manipulate.

A special structure

In the new work, a Los Alamos team led by Scott Crooker focused on Co_1/3NbS₂ because of its topological nature. In this material, layers of cobalt atoms are positioned, or intercalated, between monolayers of niobium disulfide, creating 2D triangular lattices with ABAB stacking. The spins of these cobalt atoms point either toward or away from the centers of the tetrahedra formed by the atoms. The result is a noncoplanar spin ordering that produces a chiral, or “handed,” spin texture.

This chirality affects the motion of electrons in the material because when an electron passes through a chiral pattern of spins, it picks up a geometrical phase known as a Berry phase. This makes it move as if it were “seeing” a region with a real magnetic field, giving the material a nonzero Hall conductivity which, in turn, affects how it absorbs circularly polarized light.

Characterizing a topological antiferromagnet

To characterize this behaviour, the researchers used an optical technique called magnetic circular dichroism (MCD) that measures the difference in absorption between left and right circularly polarized light and depends explicitly on the Hall conductivity.

Similar to the MCD that is measured in well-known ferromagnets such as iron or nickel, the amplitude and sign of the MCD measured in Co_1/3NbS₂ varied as a function of the wavelength  of the light.  This dependence occurs because light prompts optical transitions between filled and empty energy bands. “In more complex materials like this, there is a whole spaghetti of bands, and one needs to consider all of them,” Crooker explains. “Precisely which mix of transitions are being excited depends of course on the photon energy, and this mix changes with energy. Sometimes the net response is positive, sometimes negative; it just depends on the details of the band structure.”

To understand the mix of transitions taking place, as well as the topological character of those transitions, scientists use the concept of Berry curvature, which is the momentum-space version of the magnetic field-like effect described earlier. If the accumulated Berry phase is positive (negative), then the electron is moving in a right-handed (left-handed) spin texture chirality, which is captured by the Berry curvature of the band structure in momentum space.

Imaging and painting chiral AFM domains

To image directly the domains with positive and negative chirality, the researchers cooled the sample below its ordering temperature, shined light of a particular wavelength onto it, and measured its MCD using a scanning MCD microscope. The sign of the measured MCD value revealed the chirality of the AFM domains.

To “write” a different chirality into these AFM domains, the researchers again cooled the sample below its ordering temperature, this time in the presence of a small positive magnetic field B, which fixed the sample in a positive chiral AFM state. They then reversed the polarity of B and illuminated a spot of the sample to heat it above the ordering temperature. Once the spot cooled down, the negative-polarity B-field changed the AFM state in the illuminated region into a negative chirality. When the “painting” was finished, the researchers imaged the patterns with the MCD microscope.

In the past, a similar thermo-magnetic scheme gave rise to ferromagnetic-based data storage disks. This work, which is published in Physical Review Letters, marks the first time that light has been used to manipulate AFM chiral domains – a fundamental requirement for developing AFM-based information storage technology and spintronics.  In the future, Crooker says the group plans to extend this technique to characterize other complex antiferromagnets with nontrivial magnetic configurations, use light to “write” interesting spatial patterns of chiral domains (patterns of Berry phase), and see how this influences electrical transport.

The post Visible light paints patterns onto chiral antiferromagnets appeared first on Physics World.

Grey, ugly, dull. Concrete is not the most exciting material in the world. That is, until you start to think about its impact on our lives. Concrete is the second most consumed material on the planet after water. Humanity uses about 30 billion tonnes of the stuff every year, the equivalent of building an entire new New York City every month. Put another way, there is so much concrete in the world and so much being made that by the 2040s it will outweigh all living matter.

As the son of a builder, I have made a few concrete mixes over the years myself, usually following my father’s tried and trusted recipe. Take one part cement (fine mineral powder), two parts sand, and four parts aggregate (crushed stone), then mix and add enough water until it all goes gloopy.

The ubiquity and low cost of these simple ingredients are just two of the reasons for concrete’s global reach. In liquid form, it can be moulded into almost any shape, and once set, it is as hard and durable as stone. What’s more, it doesn’t burn, rot or get eaten by animals.

These factors make concrete the ideal material for everything from vast imposing dams to sleek kitchen floors. However, its gargantuan presence across society comes at an equally epic environmental cost. If concrete were a country, it would rank third behind only the US and China as a greenhouse gas emitter.

Though raw material processing and transport of concrete are part of the problem, concrete’s biggest environmental impact comes from the heat and chemical processes involved in producing cement. Ordinary cement clinker (the raw form of cement before it is ground to a powder) is the product of heating limestone up to 1450 °C until it breaks apart into lime and carbon dioxide (CO₂). This heating requires lots of energy and the chemical process releases huge amounts of the greenhouse gas CO₂ – meaning that cement makes up around 90% of the carbon footprint of an average concrete mix.

In the UK and some other parts of the world, this climate impact is well recognized, with the industry having made significant efforts to decarbonize over the last few decades. “Since 1990, the UK concrete industry has decreased its direct and indirect environmental impacts by over 53% through various technology levers,” says Elaine Toogood – an architect and senior director at the Mineral Products Association’s Concrete Centre, the UK’s technical hub for all things concrete.

This reduction has been achieved through actions such as fuel switching, decarbonizing electricity and transport networks, and carbon capture technology. “For example, over 50% of all the heat that’s needed to make cement is now supplied by waste-derived fuels,” Toogood adds.

Yet the sheer scale of the global concrete industry means that much more needs to be done to fully mitigate concrete’s carbon impact. Can physics, and more specifically AI, lend a hand?

Low-carbon replacements

Replacing cement – concrete’s least green ingredient – with low-carbon alternatives seems like a good place to start. Two well-proven options have been available for decades.

Fly ash – the by-product of burning coal at power plants – can replace about 30% of cement in concrete mixes. It has been used in the construction of many prominent structures including the Channel Tunnel, which opened in 1994. Blast furnace slag – the by-product of iron and steel production – is another capable replacement, and can make up to 70% of cement content. Slag was used in 2009 to substitute half of the regular cement in the precast concrete units that now make up the sea defences on Blackpool beach.

Yet although these waste materials are currently extensively used as cement or concrete additions in the UK and elsewhere, they rely on very polluting sources (coal-fired power plants and blast furnaces) that are gradually being phased out globally to meet climate targets. As a result, fly ash and blast furnace slag are not long-term solutions. New low-carbon materials are needed, which is where physics can play a decisive role.

Based in Debre Tabor University in Ethiopia, Gashaw Abebaw Adanu is an expert in innovative construction materials. In 2021 he and colleagues investigated the potential of partially replacing (0%, 5%, 10%, 15% and 20%) standard cement with ash from burning lowland Ethiopian bamboo leaf, a common local construction waste material (Adv. Civ. Eng 10.1155/2021/6468444). The findings were encouraging. Though the concrete took longer to set with increased bamboo leaf ash content, the material’s strength, water absorption and sulphate attack (concrete breakdown caused by sulphate ions reacting with the hardened cement paste) improved for 5–10% bamboo leaf ash mixes. The results suggest that up to 10% of cement could be swapped for this local low-carbon alternative.

Steel, copper – or hair?

More recently, Adanu has turned his focus to concrete fibre reinforcement. Adding small amounts of steel, copper or polyethene fibres is known to increase concrete’s ductility and crack resistance by up to 200% and 90%, respectively. The tiny fibres act like micro-stitches throughout the entire mix, transforming concrete from a brittle material into a tough, energy-absorbing composite.

Fibre reinforcement also leads to major cost savings and a reduced carbon footprint, primarily by removing the need for traditional steel rebar and mesh, where 50 kg of steel fibres can often do the work of 100 kg+ of traditional rebar. Eliminating this expensive material also reduces labour and maintenance costs.

In his latest research, Adanu has explored an unexpected alternative fibre reinforcement material that would decrease costs further as it would otherwise go to landfill: human hair (Eng. Res. Express 7 015115). Adanu took waste hair from barbershops in Debre Tabor (with permission, of course), and added small amounts of it in different quantities to standard concrete mixes. “It’s not biodegradable, it’s not compostable, but as a fibre reinforcement material, we found that using 1–2% human hair improves the concrete’s tensile strength, compressive strength, cracking resistance and reduces shrinkage,” says Adanu. “It makes concrete more clean and sustainable, and because it improves the quality of the concrete, it reduces cost at the same time.”

Research like Adanu’s, involving experimentation with local materials, has been the driving force for innovation in construction for millennia. From the ancient Neolithic practice of boosting mudbricks’ strength by adding local straw, to the Romans using volcanic dust as high-quality cement for concrete constructions like the Pantheon in Rome – a structure that still stands to this day, with its 43.3-m diameter non-reinforced concrete dome remaining the largest in the world. But testing one material at a time is no longer the only way.

Taking a more modern, wide-ranging approach, a team of researchers led by Soroush Mahjoubi and Elsa Olivetti of Massachusetts Institute of Technology (MIT), recently mined the cement and concrete literature, and a database of over one million rock samples, looking for cement ingredient substitutes (Communications Materials 6 99). The study not only confirmed the potential of the well-known alternatives fly ash and metallurgic slags, but also various biomass ashes like the bamboo leaf ash Adanu investigated, as well as rice husk, sugarcane bagasse, wood, tree bark and palm oil fuel ashes.

The meta-review in addition identified various other waste materials with high potential. These include construction and demolition wastes (ceramics, bricks, concrete), waste glass, municipal solid waste incineration ashes, and mine tailings (iron ore, copper, zinc), as well as 25 igneous rock types that could significantly reduce cement’s carbon impact.

AI to the rescue

Although a number of these alternative concrete materials have been known for some time, they have struggled to make an impact, with very few being used to partially replace regular cement in ready-mix concretes. Getting construction companies or concrete contractors to give them a try is no simple task.

“Concrete contractors are used to using certain mixes for certain jobs at certain times of the year, so they can plan a site and project based on how those materials are going to behave,” says Toogood. “Newer mixes act slightly differently when fresh,” she adds, which makes life tricky for those running a construction site, where concrete that behaves in a predictable manner is critical so that things run smoothly and efficiently.

Two physicists – Raphael Scheps and Gideon Farrell – aim to build this trust in low-carbon alternatives through their UK construction technology company Converge. Starting out using sensors to measure the real-time performance of different mixes of concrete in situ, they have built one of the world’s largest datasets on the performance of concrete.

They can now apply an AI model underpinned by physics principles. The program simulates the physical and chemical interactions of different components to predict the performance of a vast number of concrete mixes in a wide range of situations to a high level of accuracy. And this is key, as it builds trust to experiment with lower-carbon mixes. “With projects in the UK and Australia, we’ve helped people tweak the mix that they’re using and achieve quite major carbon savings,” says Scheps. “Anywhere from 10% all the way up to 44%.”

Currently used to recommend existing cost-saving concrete recipes, Scheps sees Converge’s AI model becoming more sophisticated over time. “As it starts to uncover the real fundamental physics-based rules for what drives concrete chemistry, our model will make projections for entirely new materials,” he enthuses.

Also exploring the power of AI to optimize concrete production is US company Concrete.ai. Like Converge, Concrete.ai was born from the idea of applying physics principles to optimize traditional materials and industries; specifically, how AI can be used to reduce the carbon footprint of concrete. And also like Converge, the company’s technology rests on one of the world’s largest concrete databases, consisting of vast amounts of different recipes and materials, alongside their associated performances.

Trained on this dataset, Concrete.ai’s generative AI model creates millions of possible mix designs to identify the optimal concrete recipe for any particular application. “The main difference between a solution like Concrete.ai’s and general models like ChatGPT or Gemini is that our goal is really to create recipes that don’t exist yet,” explains chief technology officer and co-founder Mathieu Bauchy. “Popular large language models regurgitate what they have been trained on and tend to hallucinate, whereas our model discovers new recipes that have never been produced before without breaking the laws of physics or chemistry, and in a reliable way.”

Bauchy sees Concrete.ai’s role as a bridge between concrete producers keen to cut their costs and carbon footprint, and innovators like Adanu or the MIT group exploring new low-carbon concrete materials who are unable to demonstrate the performance of these materials in real-world scenarios and at scale.

Circular benefits

It is perhaps apt that the industry most in need of AI insights from the likes of Converge, Concrete.ai and their growing number of competitors is the AI industry itself. New data centres being used to train, deploy and deliver AI applications and services are the cause of a huge spike in the greenhouse gas emissions of tech giants such as Google, Meta, Microsoft and Amazon. And one of the biggest contributors to those emissions is the concrete from which these hyperscale facilities are built.

This is the reason Meta recently partnered with concrete maker Amrize to develop AI-optimized concrete. For Meta’s new 66,500 m² data centre in Rosemount, Minnesota, the partners applied Meta’s AI models and Amrize’s materials-engineering expertise to deliver concrete that met key criteria including high strength and low carbon content, as well as practical performance characteristics like decent cure speed and surface quality. The partners estimate that the custom mix will reduce the total carbon footprint of this concrete by 35%.

“There is an interesting synergy between concrete and AI,” says Bauchy. “AI can help design greener concrete, and on the other hand, concrete can be used to build more sustainable data centres to power AI.” With other tech giants exploring AI’s potential in reducing the carbon footprint of the concrete they use too, it may well be that the very places in which AI is developed become the testbeds for AI-derived sustainable green concrete solutions.

The post Green concrete: paving the way for sustainable structures appeared first on Physics World.

Personalized health – the use of individualized measurements to address each patient’s specific needs – is a research field that’s evolving at pace. Bringing this level of personalization into the clinic is an interdisciplinary challenge, requiring the development of sensors that generate clinically meaningful data outside the hospital, new imaging modalities and analysis techniques, and computational tools that address the uncertainties of dealing with just one individual.

Much of the most impactful work in this field sits in the spaces between established disciplines. And for researchers looking to publish their findings or read about the latest breakthroughs, this work is often scattered across discipline-specific journals. A new open access journal from IOP Publishing – Medical Sensors & Imaging (MSI) – aims to remedy this shortfall, providing a dedicated home for authors working across sensing, imaging, modelling and data-driven healthcare.

“We want a journal where physicists, engineers, computer scientists, biomedical researchers and clinicians can publish and read work that advances personalized health, without confinement into traditional silos,” explains founding editor-in-chief Marco Palombo from Cardiff University. “MSI also aims to play an important role in strengthening interdisciplinary exchange.”

“The community needs a specialized forum that doesn’t just report on new materials or a clinical trial, but validates innovations that can specifically solve complex biomedical challenges,” adds deputy editor Xiliang Luo from Qingdao University of Science and Technology. “I think this journal is a perfect fit for that gap.”

Connecting communities

Published by IOP Publishing on behalf of the Institute of Physics and Engineering in Medicine (IPEM), MSI aims to dismantle the barriers between engineering innovation and clinical application by creating a community of experts that work together to translate innovative technology into clinical settings.

MSI sits within IPEM’s journal portfolio that includes Physics in Medicine & Biology, Physiological Measurement and Medical Engineering & Physics. Its aims and scope were designed to complement, rather than overlap with, these existing journals and provide a dedicated venue for translational work and practical applied research that may otherwise struggle to fit a traditional scope.

Being part of this established family of journals brings with it strong editorial standards, an established readership base and a commitment to scientific integrity. The journal also offers rapid, high-quality peer review, with feedback that’s constructive, rigorous and fair. MSI is fully open access, which maximizes the visibility, reach and impact of its published papers.

“For a new journal in a dynamic field, ensuring content is discoverable and barrier-free is essential for building an audience quickly and establishing credibility,” says Palombo. “We also wanted MSI to support global participation. Many excellent groups operate with limited budgets but make major scientific contributions. Open access reduces inequities in who can read and build on published work.”

“For the authors, we can provide a specialized platform for scientists whose work transcends traditional boundaries, offering visibility to a broad audience that’s eager for translational solutions,” says Luo. “And for the readers, I think we will be the go-to resource for academic researchers, industry R&D leaders, and healthcare innovators seeking the latest breakthroughs in personalized health monitoring and advanced diagnostics.”

Hot topics

Palombo contributed to the strategic development of the journal at an early stage, drawing upon his experience in healthcare and medical imaging research and engaging with the research community to identify the scientific niche that MSI could fill. Working with IOP Publishing, he helped shape the journal’s aims and scope and assembled a diverse, internationally recognized editorial board with knowledge aligned with the journal’s mission – including Lui, who brings specialist expertise in wearable technologies and biosensors.

The journal will publish high-quality research on novel biomedical sensing and imaging techniques, along with the algorithms, validation frameworks and translational studies that demonstrate their application in real-world medicine. MSI also provides a platform to showcase research on hot topics such as wearable and implantable sensors for continuous physiological monitoring, for example, or microneedle-based sensing technologies and breath analysis.

The development of flexible and biocompatible materials will be key for the growth of bio-integrated devices and biodegradable or transient electronics, as will anti-fouling strategies that enable use of sensors in complex biological environments. On the imaging side, the journal scope encompasses mainstay medical imaging techniques such as MRI, CT, ultrasound, PET and SPECT, as well as emerging multimodal and hybrid approaches, with a focus on technical innovation and translational relevance.

“Given my own background, I’m particularly keen to see strong submissions in the area of MRI, including advanced quantitative biomarkers and approaches that probe tissue microstructure,” notes Palombo. “I also see huge potential in connecting imaging to computational modelling – particularly digital twins – and in building imaging pipelines that enable personalized diagnosis and prognosis.”

“Other exciting areas include combining sensing and imaging technologies into one system, and closed-loop ‘sense then act’ systems, which sense something and can then release medicine to treat the disease,” says Luo.

The rise of AI

Artificial intelligence (AI) is becoming increasingly central to both sensing and imaging, and will likely play a major role in the evolution of personalized health, enabling a shift towards multimodal fusion of sensor streams, imaging and clinical data. AI could also facilitate the introduction of integrated sensor systems that collect data and interpret signals in real time, and digital twins that link patient-specific data with computational models to simulate disease progression or treatment response.

Palombo emphasizes the importance of trustworthy AI: methods that don’t just provide an output, but are explainable, robust and explicitly handle uncertainty. This is a direction seen in the general field of AI, but is especially important within healthcare. He also cites the increasing momentum around green healthcare and green AI, with personalized health technologies designed to reduce waste and minimize energy consumption, and clinical models developed with far greater computational efficiency.

“It would be fantastic to have an AI model running directly on the sensor, for example, and this ties in with the environmental impact of AI,” he explains. “If we keep AI small and manageable, then it pollutes less, is more affordable for everybody and can be deployed on small, lightweight devices.”

A community focal point

Looking ahead, Palombo hopes that MSI will becomes a leading platform for interdisciplinary innovation in personalized health, and the routine home for publishing major advances in sensing, imaging, modelling and trustworthy AI. “Over time, I’d like the journal to build depth in core areas, while also actively shaping emerging directions such as digital twins, uncertainty-aware and explainable AI, multimodal integration and technologies that are genuinely deployable in clinical workflows.”

“Currently, the fields of sensor engineering and clinical medicine often run on parallel tracks. My hope is that this journal will force these tracks to converge over time,” adds Luo. “I see the journal fostering a new language where chemists, physicists, engineers and doctors can understand each other by publishing papers in MSI.”

• Medical Sensors & Imaging is fully open for submissions, with the first issue expected to publish in Q2/Q3 of this year. During the launch phase, IOPP is covering the article processing charge (APC) for all accepted papers, enabling early contributors to publish at no cost while helping the journal establish a strong foundation of high-quality inaugural content. Beyond this period, many authors will benefit from support through IOPP’s transformative agreements, while others may be eligible for APC waivers and discounts.

 

The post New journal aims to advance the interdisciplinary field of personalized health appeared first on Physics World.

Here at Physics World we are always on the look out for physicists with extraordinary talents outside of science. In 2023, for example we were in awe of Harvard University’s Jenny Hoffman who ran across the US in 47 days, 12 hours and 35 minutes – shattering the previous record by one week.

Now, coverage of the Winter Olympics in Italy has revealed that the Chinese freestyle skier Eileen Gu had studied physics at Stanford University. The most decorated female Olympic freestyle skier in history, US-born Gu bagged two gold medals and a silver at the 2022 Beijing games and added three silvers at Milano Cortina.

Gu has subsequently switched majors to international relations at Stanford, but we can still celebrate her as an honorary physicist.

Physics-rich event

Indeed, freestyle skiing is quite possibly the most physics-rich of all Olympic events. Athletes must consider friction, gravity and the conservation of momentum and angular momentum to perfect their skiing.

Now, I’m not suggesting that studying free-body diagrams of freestyle manoeuvres is essential for Olympic success, but I live in hope that an understanding of classical mechanics can improve one’s skiing. (I’m not sure why I believe this, because a PhD and decades of writing about physics certainly hasn’t improved my skiing!).

As well as being lauded for her prowess on the snow, Gu has found herself at the centre of an international furore regarding her choice of competing for China rather than for the US. So, international relations combined with physics seems like a very good course of study!

• Article has been updated to include Gu’s third silver medal at Milano Cortina.

The post Olympian Eileen Gu rules the piste with physics and international relations appeared first on Physics World.

A new way of extracting energy from ocean waves has been proposed by a researcher in Japan. The system couples a gyroscope to an electrical generator and could be fine tuned to extract energy from a wide range of wave conditions. A prototype of the design is currently being built for testing in a wave tank. If successful, the system could be used to generate electricity onboard ships.

Ocean waves contain huge amounts of energy and humans have tried to harness this energy for centuries. But, despite the development of myriad technologies and a number of trials, the widespread commercial conversion of wave energy remains an elusive goal. One important problem is that most generation schemes only work within a narrow range of wave conditions – and the ocean can be a very messy place.

Now, Takahito Iida at the University of Osaka has proposed a new energy-harvesting technology that uses gyroscopic flywheel system that can be tuned to absorb energy efficiently over a broad range of wave frequencies.

“Wave energy devices often struggle because ocean conditions are constantly changing,” says Iida. “However, a gyroscopic system can be controlled in a way that maintains high energy absorption, even as wave frequencies vary.”

Wobbling top

At the heart of the technology is gyroscopic precession, whereby a torque on a rotating object causes the object’s axis of rotation to trace out a circle. This is familiar to anyone who has played with a spinning top, which will wobble (precess) when perturbed.

Iida’s device is called a gyroscopic wave energy converter and comprises a spinning flywheel mounted on a floating platform (see figure). On calm seas, the gyroscope’s axis of rotation points in a fixed direction thanks to the conservation of angular momentum. However, waves will cause the platform to pitch from side-to-side, exerting torques on the gyroscope and causing it to precess.  It is this precession that drives a generator to deliver electrical power.

To design the system, Iida used linear wave theory to model the coupled interactions between waves, the platform, the gyroscope and the generator. This allowed him to devise a scheme for tuning the gyroscope frequency and generator parameters so that an energy conversion efficiency of 50% is achieved for a variety of wave conditions.

The effect of the generator was modelled as a spring-damper. This is a system that responds to a torque by storing and then returning some energy to the gyroscope (the spring), and removing some energy by converting it to electricity (the damper).  Iida discovered that a maximum conversion of 50% occurs when the spring coefficient of the generator is adjusted such that the gyroscope’s resonant frequency matches the resonant frequency of the floating platform.

Fundamental constraint

Iida explains that 50% is the maximum efficiency that can be achieved. “This efficiency limit is a fundamental constraint in wave energy theory. What is exciting is that we now know that it can be reached across broadband frequencies, not just at a single resonant condition.”

Iida tells Physics World that a small prototype (approximately 50 cm³ in size) is being built and will be tested in a 100 m-long tank.

The next step will be the development of a system with a generating capacity of about 5 kW. Iida says that the ultimate goal is a 300 kW generator.

Iida also explains that the gyroscopic wave energy converter is designed to operate untethered to the seabed. As a result he says the system would be ideal for use as an auxiliary power system for a ship. “The target output of 300 kW is based on the assumed auxiliary power demand of a typical commercial vessel,” says Iida.

The research is described in the Journal of Fluid Mechanics.

The post Wobbling gyroscopes could harvest energy from ocean waves appeared first on Physics World.

A team headed up at TU Wien in Austria has set the Guinness World Record for creating the world’s smallest QR code. Working with industry partner Cerabyte, the researchers produced a stable and repeatedly readable QR code with an area of just 1.977 µm². When read out – using an electron microscope, as its structure is too fine to be seen with a standard optical microscope – the QR code links to a scientific webpage at TU Wien.

But this wasn’t just a ploy to get into the record books, the QR code was created as part of the team’s research into ceramic data storage materials. Unlike conventional magnetic or electronic data storage media, which degrade within decades, ceramic-based storage is designed to withstand extreme temperatures, radiation, chemical corrosion and mechanical damage.

As such, information stored in ceramic materials could endure for centuries, or even millennia. And in contrast to today’s data centres, ceramics preserve stored information without any energy input and without requiring cooling.

To create these ultralong-life data storage systems, the researchers use focused ion beams to mill the QR code into a thin film of chromium nitride, a durable ceramic often used to coat high-performance cutting tools. As each individual pixel is just 49 nm in size, roughly 10 times smaller than the wavelength of visible light, the code cannot be imaged using visible light. But when examined with an electron microscope, the QR code could indeed be read out reliably.

After the writing process, the entire stack of ceramic films is subjected to extreme conditions, such as high temperatures, corrosive environments and mechanical stress, to evaluate the material’s long-term durability and readout stability.

Pushing storage to its limits

Creating a “tiny QR code” was not the team’s initial goal, but emerged as a natural outcome of pushing this storage technology to its limits, says Paul Mayrhofer from TU Wien’s Institute of Materials Science and Technology.

“During a discussion with one of my PhD students, Erwin Peck, we realised that the writing procedure we had developed already produced features smaller than what had previously been reported for QR codes,” he explains. “This sparked the idea: if we can reliably write structures at that scale, why not intentionally create the smallest QR code possible?”

To claim its place in the record books, the QR code was successfully milled and read out in the presence of witnesses and its size independently verified using calibrated scanning electron microscopy at the University of Vienna. It is now officially recognized by Guinness as the world’s smallest QR code, and is roughly one third the size of the previous record holder.

Mayrhofer points out that the storage capacity of the ceramic data storage technology far surpasses that of a single QR code. “Based on current estimates, a cartridge of 100 x 100 x 20 mm with ceramic storage medium could potentially store on the order of 290 terabytes of raw data,” he says.

As well as offering this impressive raw capacity, for practical applications it’s also crucial that the ceramic storage offers high writing speed, which determines how efficiently large datasets can be stored, and low energy consumption during writing, which will influence the potential for scalability and sustainability. The researchers are currently working to optimize both of these parameters.

“Humanity has preserved information for millennia when carved in stone, yet much of today’s digital information risks being lost within decades,” project leader Alexander Kirnbauer tells Physics World. “Our long-term goal is to create an ultrastable, sustainable data storage technology capable of preserving information for extremely long times – potentially thousands to millions of years. In essence, we want to develop a form of storage that ensures the knowledge of our digital age does not disappear over time.”

The post World’s smallest QR code paves the way for ultralong-life data storage appeared first on Physics World.

Developing practical technologies for quantum information systems requires the cooperation of academic researchers, national laboratories and industry. That is the mission of the  Quantum Systems Accelerator (QSA), which is based at the Lawrence Berkeley National Laboratory in the US.

The QSA’s director Bert de Jong is my guest in this episode of the Physics World Weekly podcast. His academic research focuses on computational chemistry and he explains how this led him to realise that quantum phenomena can be used to develop technologies for solving scientific problems.

In our conversation, de Jong explains why the QSA is developing a range of  qubit platforms − including neutral atoms, trapped ions, and superconducting qubits – rather than focusing on a single architecture. He champions the co-development of quantum hardware and software to ensure that quantum computing is effective at solving a wide range of problems from particle physics to chemistry.

We also chat about the QSA’s strong links to industry and de Jong reveals his wish list of scientific problems that he would solve if he had access today to a powerful quantum computer.

The post Quantum Systems Accelerator focuses on technologies for computing appeared first on Physics World.

A newly identified metallic material that conducts heat nearly three times better than copper could redefine thermal management in electronics. The material, which is known as theta-phase tantalum nitride (θ-TaN), has a thermal conductivity comparable to low-grade diamond, and its discoverers at the University of California Los Angeles (UCLA), US say it breaks a record on heat transport in metals that had held for more than 100 years.

Semiconductors and insulators mainly carry heat via vibrations, or phonons, in their crystalline lattices. A notable example is boron arsenide, a semiconductor that the UCLA researchers previously identified as also having a high thermal conductivity. Conventional metals, in contrast, mainly transport heat via the flow of electrons, which are strongly scattered by lattice vibrations.

Heat transport in θ-TaN combines aspects of both mechanisms. Although the material retains a metal-like electronic structure, study leader Yongjie Hu explains that its heat transport is phonon-dominated. Hu and his UCLA colleagues attribute this behaviour to the material’s unusual crystal structure, which features tantalum atoms interspersed with nitrogen atoms in a hexagonal pattern. Such an arrangement suppresses both electron–phonon and phonon–phonon scattering, they say.

Century-old upper limit for metallic heat transport

Materials with high thermal conductivity are vital in electronic devices because they remove excess heat that would otherwise impair the devices’ performance. Among metals, copper has long been the material of choice for thermal management thanks to its relative abundance and its thermal conductivity of around 400 Wm⁻¹ K⁻¹, which is higher than any other pure metal apart from silver.

Recent theoretical studies, however, had suggested that some metallic-like materials could break this record. θ-TaN, a metastable transition metal nitride, was among the most promising contenders, but it proved hard to study because high-quality samples were previously unavailable.

Highest thermal conductivity reported for a metallic material to date

Hu and colleagues overcame this problem using a flux-assisted metathesis reaction. This technique removed the need for the high pressures and temperatures required to make pure samples of the material using conventional techniques.

The team’s high-resolution structural measurements revealed that the as-synthesized θ-TaN crystals had smooth, clean surfaces and ranged in size from 10 to 100 μm. The researchers also used a variety of techniques, including electron diffraction, Raman spectroscopy, single-crystal X-ray diffraction, high-resolution transmission electron microscopy and electron energy loss spectroscopy to confirm that the samples contained single crystals.

The researchers then turned their attention to measuring the thermal conductivity of the θ-TaN crystals. They did this using an ultrafast optical pump-probe technique based on time-domain thermoreflectance, a standard approach that had already been used to measure the thermal conductivity of high-thermal-conductivity materials such as diamond, boron phosphide, boron nitride and metals.

Hu and colleagues made their measurements at temperatures between 150 and 600 K. At room temperature, the thermal conductivity of the θ-TaN crystals was 1100 Wm⁻¹ K⁻¹. “This represents the highest value reported for any metallic materials to date,” Hu says.

The researchers also found that the thermal conductivity remained uniformly high across an entire crystal. H says this reflects the samples’ high crystallinity, and it also confirms that the measured ultrahigh thermal conductivity originates from intrinsic lattice behaviour, in agreement with first-principles predictions.

Another interesting finding is that while θ-TaN has a metallic electronic structure, its thermal conductivity decreased with increasing temperature. This behaviour contrasts with the weak temperature dependence typically observed in conventional metals, in which heat transport is dominated by electrons and is limited by electron-phonon interactions.

Applications in technologies limited by heat

As well as cooling microelectronics, the researchers say the discovery could have applications in other technologies that are increasingly limited by heat. These include AI data centres, aerospace systems and emerging quantum platforms.

The UCLA team, which reports its work in Science, now plans to explore scalable ways of integrating θ-TaN into device-relevant platforms, including thin films and interfaces for microelectronics. They also aim to identify other candidate materials with lattice and electronic dynamics that could allow for similarly efficient heat transport.

The post Metallic material breaks 100-year thermal conductivity record appeared first on Physics World.

Synthetic materials such as plastics are designed to be durable and water resistant. But the processing required to achieve these properties results in a lack of biodegradability, leading to an accumulation of plastic pollution that affects both the environment and human health. Researchers at the Institute for Bioengineering of Catalonia (IBEC) are developing a possible replacement for plastics: a novel biomaterial based on chitin, the second most abundant natural polymer on Earth.

“Every year, nature produces on the order of 10¹¹ tonnes of chitin, roughly equivalent to more than three centuries of today’s global plastic production,” says study leader Javier G Fernández. “Chitin and [its derivative] chitosan are the ultimate natural engineering polymers. In nature, variations of this material produce stiff insect wings enabling flight, elastic joints enabling extraordinary jumping in grasshoppers, and armour-like protective exoskeletons in lobsters or clams.”

But while biomaterials provide a more environmentally friendly alternative to conventional plastics, most biological materials weaken when exposed to water. In this latest work, Fernández and first author Akshayakumar Kompa took inspiration from nature and developed a new biomaterial that increases its strength when in contact with water, while maintaining its natural biodegradability.

Metal matters

In the exoskeletons of insects and crustaceans, chitin it is secreted in a gel-like form into water and then transitions into a hard structure. Following a chance observation that removing zinc from a sandworm’s fangs caused them to soften in water, Kompa and Fernández investigated whether adding a different transition metal, nickel, to chitosan could have the opposite effect.

By mixing nickel chloride solution (at concentrations from 0.6 to 1.4 M) with dispersions of chitosan extracted from discarded shrimp shells, the researchers entrapped varying amounts of nickel within the chitosan structure. Fourier-transform infrared spectra of resulting chitosan films revealed the presence of nickel ions, which form weak hydrogen bonds with water molecules and increase the biomaterial’s capacity to bond with water.

“In our films, water molecules form reversible bridges between polymer chains through weak interactions that can rapidly break and reform under load,” Fernández explains. “That fast reconfiguration is what gives the material high strength and toughness under wet conditions: essentially a built-in, stress-activated ‘self-rearrangement’ mechanism. Nickel ions act as stabilizing anchors for these water-mediated bridges, enabling more and longer-range connections and making inter-chain connectivity more robust”.

The nickel-doped chitosan samples had tensile strengths of between 30 and 40 MPa, similar to that of standard plastics. Adding low concentrations of nickel did not significantly impact the mechanical properties of the films. Concentrations of 1 M or more, however, preserved the material’s strength while increasing its toughness (the ability to stretch before breaking) – a key goal in the field of structural materials and a feature unique to biological composites.

Upon immersion in water, the nickel-doped films exhibited greater tensile strength, increasing from 36.12±2.21 MPa when dry to 53.01±1.68 MPa, moving into the range of higher-performance engineering plastics. In particular, samples created from an optimal 0.8 M nickel concentration almost doubled in strength when wet (and were used for the remainder of the team’s experiments).

Scaling production

The manufacturing process involves an initial immersion in water, followed by drying for 24 h and then re-wetting. During the first immersion, any nickel ions that are not incorporated into the material’s functional bridging network are released into the water, ensuring that nickel is present only where it is structurally useful.

The researchers developed a zero-waste production cycle in which this water is used as a primary component for fabricating the next object. “The expelled nickel is recovered and used to make the next batch of material, so the process operates at essentially 100% nickel utilization across batches,” says Fernández.

They used this process to produce various nickel-doped chitosan objects, including watertight containers and a 1 m² film that could support a 20 kg weight after 24 h of water immersion. They also created a 244 x 122 cm film with similar mechanical behaviour to the smaller samples, demonstrating the potential for rapid scaling to ecologically relevant scales. A standard half-life test revealed that after approximately four months buried in garden soil, half of the material had biodegraded.

The researchers suggest that the biomaterial’s first real-world use may be in sectors such as agriculture and fishing that require strong, water-compatible and ultimately biodegradable materials, likely for packaging, coatings and other water-exposed applications. Both nickel and chitosan are already employed within biomedicine, making medicine another possible target, although any new medical product will require additional regulatory and performance validation.

The team is currently setting up a 1000 m² lab facility in Barcelona, scheduled to open in 2028, for academia–industry collaborations in sustainable bioengineering research. Fernández suggests that we are moving towards a “biomaterial age”, defined by the ability to “control, integrate, and broadly use biomaterials and biological principles within engineering applications”.

“Over the last 20 years, working on bioinspired manufacturing, we have been able to produce the largest bioprinted objects in the world, demonstrated pathways for resource-secure and sustainable production in urban environments, and even explored how these approaches can support interplanetary colonization,” he tells Physics World. “Now we are achieving material properties that were considered out of reach by designing the material to work with its environment, rather than isolating itself from it.”

The researchers report their findings in Nature Communications.

The post Nickel-enhanced biomaterial becomes stronger when wet appeared first on Physics World.

Electronics made from certain atomically thin materials can survive harsh radiation environments up to 100 times longer than traditional silicon-based devices. This finding, which comes from researchers at Fudan University in Shanghai, China, could bring significant benefits for satellites and other spacecraft, which are prone to damage from intense cosmic radiation.

Cosmic radiation consists of a mixture of heavy ions and cosmic rays, which are high-energy protons, electrons and atomic nuclei. The Earth’s magnetic field protects us from 99.9% of this ionizing radiation, and our atmosphere significantly attenuates the rest. Space-based electronics, however, have no such protection, and this radiation can damage or even destroy them.

Adding radiation shielding to spacecraft mitigates these harmful effects, but the extra weight and power consumption increases the spacecraft’s costs. “This conflicts with the requirements of future spacecraft, which call for lightweight and cost-effective architectures,” says team leader Peng Zhou, a physicist in Fudan’s College of Integrated Circuits and Micro-Nano Electronics. “Implementing radiation tolerant electronic circuits is therefore an important challenge and if we can find materials that are intrinsically robust to this radiation, we could incorporate these directly into the design of onboard electronic circuits.”

Promising transition-metal dichalcogenides

Previous research had suggested that 2D materials might fit the bill, with transistors based on transition-metal dichalcogenides appearing particularly promising. Within this family of materials, 2D molybdenum disulphide (MoS₂) proved especially robust to irradiation-induced defects, and Zhou points out that its electrical, mechanical and thermal properties are also highly attractive for space applications.

The studies that revealed these advantages were, however, largely limited to simulations and ground-based experiments. This meant they were unable to fully replicate the complex and dynamic radiation fields such circuits would encounter under real space conditions.

Better than NMOS transistors

In their work, Zhou and colleagues set out to fill this gap. After growing monolayer 2D MoS₂ using chemical vapour deposition, they used this material to fabricate field-effect transistors. They then exposed these transistors to 10 Mrad of gamma-ray irradiation and looked for changes to their structure using several techniques, including cross-sectional transmission electron microscopy (TEM) imaging and corresponding energy-dispersive spectroscopy (EDS) mapping.

These measurements indicated that the 2D MoS₂ in the transistors was about 0.7 nm thick (typical for a monolayer structure) and showed no obvious signs of defects or damage. Subsequent Raman characterization on five sites within the MoS₂ film confirmed the devices’ structural integrity.

The researchers then turned their attention to the transistors’ electrical properties. They found that even after irradiation, the transistors’ on-off ratios remained ultra-high, at about 10⁸. They note that this is considerably better than a similarly-sized Si N-channel metal–oxide–semiconductor (NMOS) transistors fabricated through a CMOS process, for which the on-off ratio decreased by a factor of more than 4000 after the same 10 Mrad irradiation.

The team also found that MoS₂ system consumes only about 49.9 mW per channel, making its power requirement at least five times lower than the NMOS one. This is important owing to the strict energy limitations and stringent power budgets of spacecraft, Zhou says.

Surviving the space environment

In their final experiment, the researchers tested their MoS₂ structures on a spacecraft orbiting at an altitude of 517 km, similar to the low-Earth orbit of many communication satellites. These tests showed that the bit-error rate in data transmitted from the structures remained below 10^-8 even after nine months of operation, which Zhou says indicates significant radiation and long-term stability. Indeed, based on test data, electronic devices made from these 2D materials could operate for 271 years in geosynchronous orbit – 100 times longer than conventional silicon electronics.

“The discovery of intrinsic radiation tolerance in atomically thin 2D materials, and the successful on-orbit validation of the atomic-layer semiconductor-based spaceborne radio-frequency communication system have opened a uniquely promising pathway for space electronics leveraging 2D materials,” Zhou says. “And their exceptionally long operational lifetimes and ultra-low power consumption establishes the unique competitiveness of 2D electronic systems in frontier space missions, such as deep-space exploration, high-Earth-orbit satellites and even interplanetary communications.”

The researchers are now working to optimize these structures by employing advanced fabrication processes and circuit designs. Their goal is to improve certain key performance parameters of spaceborne radio-frequency chips employed in inter-satellite and satellite-to-ground communications. “We also plan to develop an atomic-layer semiconductor-based radiation-tolerant computing platform, providing core technological support for future orbital data centres, highly autonomous satellites and deep-space probes,” Zhou tells Physics World.

The researchers describe their work in Nature.

The post 2D materials help spacecraft electronics resist radiation damage appeared first on Physics World.

In this work, the researchers theoretically explore how quantum materials can transition continuously from one ordered state to another, for example, from a magnetic phase to a phase with crystalline or orientational order. Traditionally, such order‑to‑order transitions were thought to require fractionalisation, where particles effectively split into exotic components. Here, the team identifies a new route that avoids this complexity entirely.

Their mechanism relies on two renormalisation‑group fixed points in the system colliding and annihilating, which reshapes the flow of the system and removes the usual disordered phase. A separate critical fixed point, unaffected by this collision, then becomes the new quantum critical point linking the two ordered phases. This allows for a continuous, seamless transition without invoking fractionalised quasiparticles.

The authors show that this behaviour could occur in several real or realistic systems, including rare‑earth pyrochlore iridates, kagome quantum magnets, quantum impurity models and even certain versions of quantum chromodynamics. A striking prediction of the mechanism is a strong asymmetry in energy scales on the two sides of the transition, such as a much lower critical temperature and a smaller order parameter where the order emerges from fixed‑point annihilation.

This work reveals a previously unrecognised kind of quantum phase transition, expands the landscape beyond the usual Landau-Ginzburg-Wilson framework, which is the standard theory for phase transitions, and offers new ways to understand and test the behaviour of complex quantum systems.

Do you want to learn more about this topic?

Dynamical quantum phase transitions: a review by Markus Heyl (2018)

The post Rethinking how quantum phases change appeared first on Physics World.

Unlike crystals, whose atoms arrange themselves in tidy, repeating patterns, glass is a non‑equilibrium material. A glass is formed when a liquid is cooled so quickly that its atoms never settle into a regular pattern, instead forming a disordered, unstructured arrangement.

In this process, as temperature decreases, atoms move more and more slowly. Near a certain temperature –the glass transition temperature – the atoms move so slowly that the material effectively stops behaving like a liquid and becomes a glass.

This isn’t a sharp, well‑defined transition like water turning to ice. Instead, it’s a gradual slowdown: the structure appears solid long before the atoms would theoretically cease to rearrange.

This slowdown can be extrapolated and be used to predict the temperature at which the material’s internal rearrangement would take infinitely long. This hypothetical point is known as the ideal glass transition. It cannot be reached in practice, but it provides an important reference for understanding how glasses behave.

Despite years of research, it’s still not clear exactly how glass properties depend on how it was made – how fast it was cooled, how long it aged, or how it was mechanically disturbed. Each preparation route seems to give slightly different behaviour.

For decades, scientists have struggled to find a single measure that captures all these effects. How do you describe, in one number, how disordered a glass is?

Recent research has emerged that provides a compelling answer: a configurational distance metric. This is a way of measuring how far the internal structure of a piece of glass is from a well‑defined reference state.

When the researchers used this metric, they could neatly collapse data from many different experiments onto a single curve. In other words, they found a single physical parameter controlling the behaviour.

This worked across a wide range of conditions: glasses cooled at different rates, allowed to age for different times, or tested under different strengths and durations of mechanical probing.

As long as the experiments were conducted above the ideal glass transition temperature, the metric provided a unified description of how the material dissipates energy.

This insight is significant. It suggests that even though glass never fully reaches equilibrium, its behaviour is still governed by how close it is to this idealised transition point. In other words, the concept of the kinetic ideal glass transition isn’t just theoretical, it leaves a measurable imprint on real materials.

This research offers a powerful new way to understand and predict the mechanical behaviour of glasses in everyday technologies, from smartphone screens to industrial coatings.

The post How a Single Parameter Reveals the Hidden Memory of Glass appeared first on Physics World.

 

Electrochemical CO­₂ reduction converts CO­₂ to higher-value products using an electrocatalyst and could pave the way for electrification of the chemical industry. A key challenge for CO­₂ reduction is its poor selectivity (faradaic efficiency) due to competition with the hydrogen evolution reaction in aqueous electrolytes. Rotating ring-disk electrode (RRDE) experiments have become a popular method to quantify faradaic efficiencies, especially for gold electrocatalysts. However, such measurements suffer from poor inter-laboratory reproducibility. This work identifies the causes of variability in RRDE selectivity measurements by comparing protocols with different electrochemical methods, reagent purities, and glassware cleaning procedures. Electroplating of electrolyte impurities onto the disk and ring surfaces were identified as major contributors to electrocatalyst deactivation. These results highlight the need for standardized and cross-laboratory validation of CO₂RR selectivity measurements using RRDE. Researchers implementing this technique for CO₂RR selectivity measurements need to be cognizant of electrode deactivation and its potential impacts on faradaic efficiencies and overall conclusions of their work.

Maria Kelly is a Jill Hruby Postdoctoral Fellow at Sandia National Laboratories. She earned her PhD in Professor Wilson Smith’s research group at the University of Colorado Boulder and the National Renewable Energy Laboratory. Her doctoral work focused on characterization of carbon dioxide conversion interfaces using analytical electrochemical and in situ scanning probe methods. Her research interests broadly encompass advancing experimental measurement techniques to investigate the near-electrode environment during electrochemical reactions.

 

 

The post Challenges in CO₂ Reduction Selectivity Measurements by Hydrodynamic Methods appeared first on Physics World.

Pairs of nonidentical particles trapped in adjacent nodes of a standing wave can harvest energy from the wave and spontaneously begin to oscillate, researchers in the US have shown. What is more, these interactions appear to violate Newton’s third law. The researchers believe their system, which is a simple example of a classical time crystal, could offer an easy way to measure mass with high precision. It might also, they hope, provide insights into emergent periodic phenomena in nature.

Acoustic tweezers use sound waves to create a potential-energy well that can hold an object in place – they are the acoustic analogue of optical tweezers. In the case of a single trapped object, this can be treated as a dissipationless process, in which the particle neither gains nor loses energy from the trapping wave.

In the new work, David Grier of New York University, together with graduate student Mia Morrell and undergraduate Leela Elliott, created an ultrasound standing wave in a cavity and levitated two objects (beads) in adjacent nodes.

“Ordinarily, you’d say ‘OK, they’re just going to sit there quietly and do nothing’,” says Grier; “And if the particles are identical, that’s exactly what’s going to happen.”

Breaking the law

If the two particles differ in size, material or any other property that affects acoustic scattering, they can spontaneously begin to oscillate. Even more surprisingly, this motion appears unconstrained by the requirement that momentum be conserved – Newton’s third law.

“Who ordered that?”, muses Grier.

The periodic oscillation, which has a frequency parametrized only by the properties of the particles and independent of the trapping frequency, forms a very simple type of emergent active matter called a time crystal.

The trio analysed the behaviour of adjacent particles trapped in this manner using the laws of classical mechanics, and discovered an important subtlety had been missed. When identical particles are trapped in nearby nodes, they interact by scattering waves, but the interactions are equal and opposite and therefore cancel.

“The part that had never been worked out before in detail is what happens when you have two particles with different properties interacting with each other,” says Grier. “And if you put in the hard work, which Mia and Leela did, what you find is that to the first approximation there’s nothing out of the ordinary.” At the second order, however, the expansion contains a nonreciprocal term. “That opens up all sorts of opportunities for new physics, and one of the most striking and surprising outcomes is this time crystal.”

Stealing energy

This nonreciprocity arises because, if one particle is more strongly affected by the mutual scattering than the other, it can be pushed farther away from the node of the standing wave and pick up potential energy, which can then be transferred through scattering to the other particle. “The unbalanced forces give the levitated particles the opportunity to steal some energy from the wave that they ordinarily wouldn’t have had access to,” explains Grier. The wave also carries away the missing momentum, resolving the apparent violation of Newton’s third law.

If it were acting in isolation, this energy input would make the oscillations unstable and throw the particles out of the nodes. However, energy is removed by viscosity: “If everything is absolutely right, the rate at which the particles consume energy exactly balances the rate at which they lose energy to viscous drag, and if you get that perfect, delicious balance, then the particles can jiggle in place forever, taking the fuel from the wave and dumping it back into the system as heat.” This can be stable indefinitely.

The researchers have filed a patent application for the use of the system to measure particle masses with microgram-scale precision from the oscillation frequency. Beyond this, they hope the phenomenon will offer insights into emergent periodic phenomena across timescales in nature: “Your neurons fire at kilohertz, but the pacemaker in your heart hopefully goes about once per second,” explains Grier.

The research is described in Physical Review Letters.

“When I read this I got somehow surprised,” says Glauber Silva of The Federal University of Alagoas in Brazil; “The whole thing of how to get energy from the surrounding fields and produce motion of the coupled particles is something that the theoretical framework of this field didn’t spot before.”

“I’ve done some work in the past, both in simulations and in optical systems that are analogous to this, where similar things happen, but not nearly as well controlled as in this particular experiment,” says Dustin Kleckner of University of California, Merced. He believes this will open up a variety of further questions: “What happens if you have more than two? What are the rules? How do we understand what’s going on and can we do more interesting things with it?” he says. 

The post Time crystal emerges in acoustic tweezers appeared first on Physics World.

A new cooling technique based on the principles of dissolution barocaloric cooling could provide an environmentally friendly alternative to existing refrigeration methods. With a cooling capacity of 67 J/g and an efficiency of nearly 77%, the method developed by researchers from the Institute of Metal Research of the Chinese Academy of Sciences can reduce the temperature of a sample by 27 K in just 20 seconds – far more than is possible with standard barocaloric materials.

Traditional refrigeration relies on vapour-compression cooling. This technology has been around since the 19^th century, and it relies on a fluid changing phase. Typically, an expansion valve allows a liquid refrigerant to evaporate into a gas, absorbing heat from its surroundings as it does so. A compressor then forces the refrigerant back into the liquid state, releasing the heat.

While this process is effective, it consumes a lot of electricity, and there is not much room for improvement. After more than a century of improvements, the vapour-compression cycle is fast approaching the maximum efficiency set by the Carnot limit. The refrigerants are also often toxic, contributing to environmental damage.

In recent years, researchers have been exploring caloric cooling as a possible alternative. Caloric cooling works by controlling the entropy, or disorder, within a material using magnetic or electric fields, mechanical forces or applied pressure. The last option, known as barocaloric cooling, is in some ways the most promising. However, most of the known barocaloric materials are solids, which suffer from poor heat transfer efficiency and limited cooling capacity. Transferring heat in and out of such materials is therefore slow.

A liquid system

The new technique overcomes this limitation thanks to a fundamental thermodynamic process called endothermic dissolution. The principle of endothermic dissolution is that when a salt dissolves in a solvent, some of the bonds in the solvent break. Breaking those bonds takes energy, and so the solvent cools down – sometimes dramatically.

In the new work, researchers led by metallurgist and materials scientist Bing Li discovered a way to reverse this process by applying pressure. They began by dissolving a salt, ammonium thiocyanate (NH₄SCN), in water. When they applied pressure to the resulting solution, the salt precipitated out (an exothermic process) in line with Le Chatelier’s principle, which states that when a system in chemical equilibrium is disturbed, it will adjust itself to a new equilibrium by counteracting as far as possible the effect of the change.

When they then released the pressure, the salt re-dissolved almost immediately. This highly endothermic process absorbs a massive amount of heat, causing the temperature of the solution to drop by nearly 27 K at room temperature, and by up to 54 K at higher temperatures.

A chaotropic salt

Li and colleagues did not choose NH₄SCN by chance. The material is a chaotropic agent, meaning that it disrupts hydrogen bonding, and it is highly soluble in water, which helps to maximize the amount present in the solution during that part of the cooling cycle. It also has a large enthalpy of solution, meaning that its temperature drops dramatically when it dissolves. Finally, and most importantly, it is highly sensitive to applied pressures in the range of hundreds of megapascals, which is within the capacity of conventional hydraulic systems.

Li says that he and his colleagues’ approach, which they detail in Nature, could encourage other researchers to find similar techniques that likewise do not rely on phase transitions. As for applications, he notes that because aqueous NH₄SCN barocaloric cooling works well at high temperatures, it could be suited to the demanding thermal management requirements of AI data centres. Other possibilities include air conditioning in domestic and industrial vehicles and buildings.

There are, however, some issues that need to be resolved before such cooling systems find their way onto the market. NH₄SCN and similar salts are corrosive, which could damage refrigerator components. The high pressures required in the current system could also prove damaging over the long run, Li adds.

To address these and other drawbacks, the researchers now plan to study other such near-saturated solutions at the atomic level, with a particular focus on how they respond to pressure. “Such fundamental studies are vital if we are to optimize the performance of these fluids as refrigerants,” Li tells Physics World.

The post Giant barocaloric cooling effect offers a new route to refrigeration appeared first on Physics World.

Hydrogen is considered a clean fuel because it produces water rather than carbon dioxide when burned, and it is seen as a promising route toward lower emissions. It is especially valuable for replacing fossil fuels in industrial processes that require extremely high temperatures and are difficult to electrify. Although hydrogen itself is not a greenhouse gas like carbon dioxide, methane, or nitrous oxide (gases that trap heat in the Earth’s atmosphere), it can still indirectly contribute to warming. Normally, hydroxyl radicals, which are highly reactive atmospheric molecules made of one oxygen and one hydrogen atom with an unpaired electron, break down methane into carbon dioxide and water. But when hydroxyl radicals react with hydrogen instead, fewer radicals are available to remove methane, allowing methane to persist longer in the atmosphere and increasing its warming effect.

This study examines how hydrogen leakage in hydrogen‑based energy systems could influence the planet. The researchers analysed 23 different U.S. future scenarios, including some that eliminate fossil fuels entirely. They estimated how much hydrogen might leak in each scenario, compared those leaks to the remaining carbon dioxide and methane emissions, and calculated how much additional emissions reductions and/or carbon removal would be needed to offset the warming from hydrogen under low, medium, and high leak rates, and over both short‑term and long‑term warming timescales.

They found that although hydrogen leaks do contribute to warming, their impact is much smaller than the warming from the remaining carbon dioxide and methane in all scenarios. Hydrogen’s warming effect appears much larger over a 20 year period because its short‑lived chemical interactions amplify methane and ozone quickly, even though its long‑term impact remains relatively modest. Only small increases in carbon dioxide removal or small reductions in other emissions are needed to offset the warming caused by hydrogen leaks. However, because estimates of hydrogen leakage rates vary widely in the scientific literature, improved measurement and monitoring are essential.

Do you want to learn more about this topic?

Hydrogen storage in liquid hydrogen carriers: recent activities and new trends Tolga Han Ulucan et al. (2023)

The post The hidden footprint of hydrogen appeared first on Physics World.

Machine-learning could help us use cosmic muons to peer inside large objects such as nuclear reactors. Developed by researchers in China, the technique is capable of identifying target materials such as uranium even if they are coated with other materials.

The muon is a subatomic particle that is essentially a heavier version of the electron. Huge numbers of cosmic muons are created in Earth’s atmosphere when cosmic rays collide with gas molecules. Thousands of cosmic muons per second rain down on every square metre of Earth’s surface and these particles can penetrate tens to hundreds of metres through solid materials.

As a result, cosmic muons are used to peer inside large objects such as nuclear reactors, volcanoes and ancient pyramids. This involves placing detectors next to an object and detecting muons that have passed through or scattered within the object. Detector data are then processed using a tomography algorithm to create a 3D image of the object’s interior.

Illicit nuclear materials

Muons tend to scatter more from high-atomic-number materials, so the technique is particularly sensitive to the presence of materials such as uranium. As a result, it has been used to create systems for the detection of illicit nuclear materials hidden in freight containers.

Muon tomography is relatively straightforward when the object is of simple construction – such as a pyramid built of stone and containing voids. Producing useful images of more complex target – such as a freight container full of unknown objects – is much more difficult. The conventional computational approach is to calculate the muon-scattering physics of many different materials and combine these data with muon-tracking algorithms. This, however, tends to require huge computational resources.

Supervised machine learning has been used to reduce the computational overhead, but this requires prior knowledge of the target materials – limiting efficacy when imaging unknown and concealed materials. What is more, many materials in complex objects are coated with other materials and these coatings can affect muon scattering.

Now, Liangwen Chen at the Institute of Modern Physics of the Chinese Academy of Sciences and colleagues have used a technique called transfer learning to improve cosmic muon tomography of objects that contain coated materials. The idea of transfer learning is to begin with knowledge of the muon-scattering parameters of bare, uncoated materials and use machine learning to predict the parameters of coated materials. Chen and colleagues believe that this is the first application of transfer learning to muon tomography.

Monte Carlo simulations

The team began by creating a database describing how cosmic muons interact with representative materials with a wide range of atomic numbers. This was done by using Geant4 to do Monte Carlo simulations of how muons interact as they pass through materials. Geant4 is the most recent incarnation of the GEANT series of computer simulations, which have been used for over 50 years to design particle detectors and interpret the data that they produce.

Chen and colleagues used Geant4 to calculate how muons are scattered within nine materials ranging from magnesium (atomic number 12) to uranium (atomic number 92). These included common elements such as aluminium, copper and iron. The geometry of the scattering involves incoming cosmic muons with energies of 1 GeV and incident angles that are typical of cosmic muons. After scattering from a material target, the simulation assumes that the muons travel though two successive detectors, which measures the scattering angles. Data were generated for bare targets of the nine materials, as well as the nine materials coated with aluminium and polyethylene. Each simulation involved 500,000 muons passing through a target.

These data were then sampled using an inverse cumulative distribution function, as well as integration and interpolation. This is done to convert the data to a form that is optimal for training a neural network.

To use these data, the team created two lightweight neural-network frameworks for transfer learning: one based on fine tuning; and the other a domain-adversarial neural network. According to the team, both frameworks were able to identify correlations between muon scattering-angle distributions and different target materials. Crucially, this was the case even when the target materials were coated in aluminium or polyethylene.

Chen explains, “Transfer learning allows us to preserve the fundamental physical characteristics of muon scattering while efficiently adapting to unknown environments under shielding”.

Chen and colleagues are now trying to apply their process to more complicated scattering geometries. The also plan to include detector effects and targets made of several materials.

“By integrating simulation, physics, and data-driven learning, this research opens new pathways for applying artificial intelligence to nuclear science and security technologies,” says Chen.

The research is described in Nuclear Science and Techniques.

The post Transfer learning could help muon tomography identify illicit nuclear materials appeared first on Physics World.

Katie Perry studied physics at the University of Surrey in the UK, staying on there to do a PhD. While at Surrey, she worked with the nuclear physicist Daphne Jackson, who was the first female physics professor in the UK. Perry later worked in science communication – both as a science writer and in public relations.

She is currently chief executive of the Daphne Jackson Trust – a charity that supports returners to research careers after a break of at least two years for family, caring or health reasons. It offers fellowships to support people to overcome the challenges of returning, ensuring that their skills, talent, training and career promise are not lost.

What skills do you use every day in your job?

One of the most important skills is multitasking and working in an agile and flexible way. I’m often travelling to meetings, conferences and other events so I have to work wherever I am, whether it’s on a train, in a hotel or at the office. How I work reminds me of a moment I had towards the end of my physics degree when suddenly everything I’d been learning seemed to fit together; I could see both the detail and the bigger picture. It’s the same now. I have to switch quickly from one project or task to another, while keeping oversight of the overall direction and operation of the charity.

I am a strong advocate for part time and flexible working, not just for me, but for all my staff and the Daphne Jackson fellows. As a manager, a key skill is to see the person and their value – not just the hours they are working. Communication and networking skills are also vital as much of my role involves developing collaborations and working with stakeholders. I could be meeting a university vice chancellor, attending a networking reception, talking to our fellows or ensuring the trust complies with charity governance – all in one day.

What do you like best and least about your job?

I love my current role, and at the risk of sounding a little cheesy, it’s because of the trust’s amazing staff and the inspiring returners we support. The fact that I knew Daphne Jackson means that leading the organization is personal to me. I’m always blown away by how inspirational, dedicated, motivated and talented our fellows are and I love supporting them to return to successful research careers. It’s a privilege to lead the charity, helping to understand the challenges and barriers that returners face – and finding ways to overcome them.

Leading a small charity requires a broad set of skills. I enjoy the variety but it’s a challenge because you’re not so much a “chief executive officer” as a “chief everything officer”. I don’t have huge teams of people to help me with, say, human resources, finance or health and safety, which makes it struggle to do them as well as I’d like. It’s therefore important to have a good work-life balance, which is why I recently took up golf. I’ve yet to have a work meeting while out practising my swing, but one day my diary might say I’m “on a course”!

What do you know today, that you wish you knew when you were starting out in your career?

If I could go back in time, I’d tell myself – like I now tell my daughter – that it’s fine not to have a defined career path or plan. Sure, it helps to have an idea of what you want to do, but you have to live and work a little to discover what you like and – more importantly – don’t like. Careers these days are highly non-linear. Unexpected life events happen so you have to adapt, just as our Daphne Jackson fellows have done.

If someone had said to me in my 20s, when I was planning a career in science communication, that I’d be a charity chief executive I wouldn’t have believed them. But here I am running a charity founded in memory of the physicist who was such a great mentor to me during my PhD. When one door closes, a window often opens – so don’t be afraid to take set off in a new direction. It can be scary, but it’s often worth the effort.

I’d also tell my younger self to network like crazy. So many opportunities have opened up because I love speaking to people. You never know who you might meet at events or what making new connections can lead to. Finally, I wish I’d known that “impostor syndrome” will always be with you – and that it’s okay to feel that way provided you recognize it and manage it. Chances are, you may never defeat it completely.

The post Ask me anything: Katie Perry – ‘I’d tell my younger self to network like crazy’ appeared first on Physics World.

More than 250 quantum scientists have signed a “manifesto” opposing the use of quantum research for military purposes. The statement – quantum scientists for disarmament –  expresses a “deep concern” about the current geopolitical situation and “categorically rejects” the militarization of quantum research or its use in population control and surveillance. The signatories now call for an open debate about the ethical implications of quantum research.

While quantum science has the potential to improve many different areas – from sensors and medicine to computing – some are concerned about its applications for military purposes. They includes quantum key distribution and cryptographic networks for communication as well as quantum clocks and sensing for military navigation and positioning.

Marco Cattaneo from the University of Helsinki in Finland, who co-authored the manifesto, says that even the potential applications of quantum technologies in warfare can be used to militarize universities and research agendas, which he says is already happening. He notes it is not unusual for scientists to openly discuss military applications at conferences or to include such details in scientific papers.

“We are already witnessing restrictions on research collaborations with fellow quantum scientists from countries that are geopolitically opposed or ambiguous with respect to the European Union, such as Russia or China,” says Cattaneo. “When talking with our non-European colleagues, we also realized that these concerns are global and multifaceted.”

Long-term aims

The idea for a manifesto originated during a quantum-information workshop that was held in Benasque in Spain between June and July 2025.

“During a session on science policy, we realized that many of us shared the same concerns about the growing militarization of quantum science and academia,” Cattaneo recalls. “As physicists, we have a strong – and terrible – historical example that can guide our actions: the development of nuclear weapons, and the way the physics community organized to oppose them and to push for their control and abolition.”

Cattaneo says that the first goal of the manifesto is to address the militarization of quantum research, which he calls “the elephant in the room”. The document also aims to raise awareness and open a debate within the community and create a forum where concerns can be shared.

“A longer-term goal is to prevent, or at least to limit and critically address, research on quantum technologies for military purposes,” says Cattaneo. He notes that “one concrete proposal” is to push public universities and research institutes to publish a database of all projects with military goals or military funding, which, he says,  “would be a major step forward.”

Cattaneo claims the group is “not naïve” and understands that stopping the technology’s military application completely will not be possible. “Even if military uses of some quantum technologies cannot be completely stopped, we can still advocate for excluding them from public universities, for abolishing classified quantum research in public research institutions, and for creating associations and committees that review and limit the militarization of quantum technologies,” he adds.

The post Quantum scientists release ‘manifesto’ opposing the militarization of quantum research appeared first on Physics World.

India has unveiled plans to build two new optical-infrared telescopes and a dedicated solar telescope in the Himalayan desert region of Ladakh. The three new facilities, expected to cost INR 35bn (about £284m), were announced by the Indian finance minister Nirmala Sitharaman on 1 February.

First up is a 3.7 m optical-infrared telescope, which is expected to come online by 2030. It will be built near the existing 2 m Himalayan Chandra Telescope (HCT) at Hanle, about 4500 m above sea level. Astronomers use the HCT for a wide range of investigations, including stellar evolution, galaxy spectroscopy, exoplanet atmospheres and time-domain studies of supernovae, variable stars and active galactic nuclei.

“The arid and high-altitude Ladakh desert is firmly established as among the world’s most attractive sites for multiwavelength astronomy,” Annapurni Subramaniam, director of the Indian Institute of Astrophysics (IIA) in Bangalore, told Physics World. “HCT has demonstrated both site quality and opportunities for sustained and competitive science from this difficult location.”

The 3.7 m telescope is a stepping stone towards a proposed 13.7 m National Large Optical-Infrared Telescope (NLOT), which is expected to open in 2038. “NLOT is intended to address contemporary astronomy goals, working in synergy with major domestic and international facilities,” says Maheswar Gopinathan, a scientist at the IIA, which is leading all three projects.

Gopinathan says NLOT’s large collecting area will enable research on young stellar systems, brown dwarfs and exoplanets, while also allowing astronomers to detect faint sources and to rapidly follow up extreme cosmic events and gravitational wave detections.

Along with India’s upgraded Giant Metrewave Radio Telescope, a planned gravitational-wave observatory in the country and the Square Kilometre Array in Australasia and South Africa, Gopinathan says that NLOT “will usher in a new era of multimessenger and multiwavelength astronomy.”

The third telescope to be supported is the 2m National Large Solar Telescope (NLST), which will be built near Pangong Tso lake 4350 m above sea level. Also expected to come online by 2030, the NLST is an advance on India’s existing 50 cm telescope at the Udaipur Solar Observatory, which provides a spatial resolution of about 100 km. Scientists also plan to combine NLST observations with data from Aditya-L1, India’s space-based solar observatory, which launched in 2023.

“We have two key goals [with NLST],” says Dibyendu Nandi, an astrophysicist at the Indian Institute of Science Education and Research in Kolkata, “to probe small-scale perturbations that cascade into large flares or coronal mass ejections and improve our understanding of space weather drivers and how energy in localised plasma flows is channelled to sustain the ubiquitous magnetic fields.”

While bolstering India’s domestic astronomical capabilities, scientists say the Ladakh telescopes – located between observatories in Europe, the Americas, East Asia and Australia – would significantly improve global coverage of transient and variable phenomena.

The post India announces three new telescopes in the Himalayan desert appeared first on Physics World.