UC Irvine to lead use of AI in solving grand challenges below Earth’s surface

The University of California Office of the President has awarded $6 million over three years to a UC Irvine-led initiative called Geophysicist.AI, which aims to apply artificial intelligence to major geophysical challenges within Earth’s crust. The project focuses on advancing sustainable geothermal energy, underground carbon dioxide sequestration, and the safe, long-term storage of spent nuclear fuel, among other critical environmental goals.

“Tapping into Earth’s abundant but hard-to-reach deep geothermal energy and better understanding and potentially predicting induced seismicity, as well as other subsurface capabilities, will take new technologies and novel approaches, and we think AI and machine learning will help us in a substantial way,” said Geophysicist.AI lead principal investigator Mohammad Javad Abdolhosseini Qomi, UC Irvine associate professor of civil and environmental engineering and materials science and engineering.

“We intend to design Geophysicist.AI with the attributes of a skilled geophysicist, including the ability to integrate and analyze heterogeneous data and models, solve mathematical representations of coupled processes across scales, and formulate and test hypotheses to provide a deeper understanding and explanation of observed geophysical processes,” he added.

With UC Irvine researchers in the lead, the project will tap the expertise of civil and environmental engineers, geoscientists, mathematicians and computer scientists at UC campuses in Riverside, San Diego, Berkeley and Santa Cruz. Also participating will be scientists from Lawrence Livermore National Laboratory and Los Alamos National Laboratory.

“Our goal is to develop a scalable artificial intelligence ecosystem that integrates large language and physics-informed models with massive amounts of real-world data to transform geophysicists’ ability to solve the most difficult subsurface challenges,” said co-principal investigator Eric Mjolsness, UC Irvine professor of computer science. “Also, we believe that the development of Geophysicist.AI will require us to employ novel methods, so both the project and its outcome will be useful more broadly in scientific applications beyond geophysics.”

Russ Detwiler, UC Irvine associate professor of civil and environmental engineering, said that as co-principal investigator, he envisions the team using Geophysicist.AI to address two main grand challenges of geoengineering. The first is to help humanity tap into enhanced geothermal systems, which entails circulating fluid through low-permeability rock as much as 2.5 miles deep to extract heat and drive turbines. Detwiler said that this endeavor – aided by AI and machine learning – is complex due to the interplay of thermal, mechanical and chemical processes at multiple scales.

The second goal is to use AI and machine learning to help predict induced seismicity from engineering pursuits beneath Earth’s surface. The researchers think a problem of this intricacy is a good match for AI since it involves terabytes of seismic data being generated continuously.

A key source of data will be the Sanford Underground Research Facility in South Dakota. Geophysicist.AI scientists will join with counterparts at the Center for Understanding Subsurface Signals and Permeability, a U.S. Department of Energy Earthshot center managed through the Pacific Northwest National Laboratory, to gain access to this resource.

In addition, the team will take advantage of the Department of Energy’s multiphysics simulators, which run on DOE supercomputers. UC Irvine also has substantial high-performance computing capabilities and deep, interdisciplinary AI expertise that will benefit the project, Mjolsness said.

“Working with our principal investigators here at UC Irvine, scientists at the other UC campuses and collaborators at the national laboratories will enable us to generate sufficient preliminary proof-of-concept results, publish as a team and [promote] synergistic activities much like a full-blown research center,” Qomi said. “Our work over the next few years should put us in a good position to compete for future federal funding.”

For more information: University of California, Irvine
Image: Mohammad Javad Abdolhosseini Qomi (left), UC Irvine associate professor of civil and environmental engineering and materials science and engineering, is lead principal investigator on the Geophysicist.AI project, while colleagues Eric Mjolsness (center), professor of computer science, and Russ Detwiler (right), associate professor of civil and environmental engineering, are co-principal investigators.

Smart phonon control boosts efficiency in eco-friendly thermoelectric material

Researchers have significantly improved the efficiency of β-Zn₄Sb₃, a thermoelectric material that converts waste heat into electricity, without using rare or costly elements. This tellurium-free compound was studied using advanced neutron scattering techniques, revealing that tiny heat vibrations—called phonons—were being disrupted by “rattling” atoms within the crystal structure. This effect, known as phonon avoided crossing, greatly reduces heat transfer through the material, enhancing its energy-harvesting capabilities.

Thanks to this effect, the material’s thermal conductivity dropped to extremely low levels—great news for thermoelectric performance. Even better, the researchers found that the single-crystal version of this material also conducts electricity better than its polycrystalline counterpart, reaching a high power conversion efficiency of 1.4%.

These results show that smart phonon control can lead to high-performance, eco-friendly materials for converting heat into power.

In thermoelectric materials, avoided crossing refers to the interaction between propagating phonons and localized vibrational modes, where their energy dispersions repel each other rather than intersect. This phenomenon occurs under specific conditions, such as crystal symmetries or vibrational mode couplings.

However, when researchers developed the single-crystal β-Zn4Sb3, they observed an unexpected, avoided crossing, revealing unique phonon behavior that deviated from conventional thermoelectric materials.

The article explores the thermoelectric performance of single-crystalline β-Zn4Sb3, a tellurium-free material, by uncovering the microscopic mechanisms that lead to its ultralow lattice thermal conductivity (κL).

Using inelastic neutron scattering (INS), the researchers provide the first experimental observation of avoided crossing between longitudinal acoustic phonons and low-energy rattling modes. This interaction causes a significant reduction in phonon group velocity—from over 4000 m/s to about 591 m/s—and shortens phonon lifetimes to under 1 picosecond, both of which contribute to strongly suppressed heat transport.

The β-Zn4Sb3 single crystal achieves a κL of approximately 0.36 W/m·K in the 300–600 K range and a peak thermoelectric figure of merit (zT) of 1.0 at 623 K. Additionally, device-level testing shows a conversion efficiency (η) of 1.4% in a single-leg thermoelectric module—one of the highest reported for undoped Zn4Sb3.

Structural characterizations via TEM reveal a grain-boundary-free lattice with uniformly distributed moiré fringes, attributed to Zn concentration variations.

These nanoscale features further enhance phonon scattering without degrading electronic performance. Compared to polycrystalline samples, the single crystal exhibits significantly better electrical conductivity due to fewer defects and optimized carrier mobility.

“This discovery shows how heat flow can be engineered to design more efficient and sustainable energy technologies—without depending on scarce resources,” says Prof. Hsin-Jay Wu.

For more information: Advanced Science

Image: Phonon dispersion map single crystalline β-Zn4Sb3 of at 300 K, measured in the longitudinal scan along [hh0]. Credit: National Taiwan University

US scientists discover new 2D material that could be used in electrochemical energy tech

Nearly a decade ago, scientists predicted that boron atoms would bond too tightly to copper to form borophene—a promising, flexible, metallic 2D material with potential in electronics, energy, and catalysis. New research led by Rice University’s Boris Yakobson confirms this prediction, but in an unexpected way, offering fresh insights into the elusive material. Yakobson emphasized that since borophene remains on the edge of existence, every new discovery about it significantly advances our understanding of materials science, physics, and electronics.

“Our very first theoretical analysis warned that on copper, boron would bond too strongly, and even if borophene did form, it would be hopelessly attached to the substrate. Now, more than a decade later, it turns out we were right ⎯ and the result is not borophene, but something else entirely,” said Yakobson.

Researchers revealed that unlike systems such as graphene on copper, where atoms may diffuse into the substrate without forming a distinct alloy, the boron atoms in this case formed a defined 2D copper boride ⎯ a new compound with a distinct atomic structure. The finding sets the stage for further exploration of a relatively untapped class of 2D materials, according to researchers.

The research reveals that since the first realization of borophene on Ag(111), two-dimensional (2D) boron nanomaterials have attracted substantial interest because of their polymorphic diversity and potential for hosting solid-state quantum phenomena.

“Here, we use atomic-resolution scanning tunneling microscopy (STM) and field-emission resonance (FER) spectroscopy to elucidate the structure and properties of atomically thin boron phases grown on Cu(111). Specifically, FER spectroscopy reveals charge transfer and electronic states that strongly differ from the decoupled borophene phases observed on silver, suggesting that the deposition of boron on copper results in strong covalent bonding characteristic of a 2D copper boride,” said researchers.

Earlier, studies synthesized borophene on metals like silver and gold, but copper remained an open ⎯ and contested ⎯ case. Some studies also highlighted that the boron might form polymorphic borophene on copper, while others suggested it could phase-separate into borides or even nucleate into bulk crystals.

Researchers revealed that resolving these possibilities required a uniquely detailed investigation combining high-resolution imaging, spectroscopy and theoretical modeling.

Yakobson underlined that what experimentalists first saw were rich patterns of atomic resolution images and spectroscopy signatures, which required a lot of hard work of interpretation.

These efforts revealed a periodic zigzag superstructure and distinct electronic signatures, both of which deviated significantly from known borophene phases. A strong match between experimental data and theoretical simulations helped resolve a debate about the nature of the material that forms at the interface between the copper substrate and the near-vacuum environment of the growth chamber, according to a press release.

“2D copper boride is likely to be just one of many 2D metal borides that can be experimentally realized,” said Mark Hersam, Walter P. Murphy Professor of Materials Science and Engineering at Northwestern University and a co-corresponding author on the study.

“We look forward to exploring this new family of 2D materials that have broad potential use in applications ranging from electrochemical energy storage to quantum information technology.”

For more information: Science Advances

Image: Calculated charge redistribution in copper boride (Cu8B14), where teal represents charge depletion and yellow represents charge accumulation.

New spectroscopy technique extends device lifespan

High-resolution, full-color display devices like foldable smartphones and ultrathin televisions use organic light-emitting diodes (OLEDs) due to their flexibility, self-illumination, lightweight construction, ultra-thin profiles, high contrast ratios, and low-voltage operation. An OLED consists of multiple ultrathin organic film layers between two electrodes, each layer playing a specific role in the device’s operation. When voltage is applied, electrical charges accumulate and emit light at the interfaces between these layers. While this multilayer structure allows precise control of charge accumulation, transport, and light generation, it also leads to degradation of the organic layers over time, limiting the lifespan and efficiency of OLED devices.

Understanding how the electronic structure at these interfaces behaves during operation remains a significant challenge. To tackle this issue, Professor Takayuki Miyamae, together with Mr. Tatsuya Kaburagi and Dr. Kazunori Morimoto from Chiba University in Japan, employed a second-order nonlinear spectroscopic method known as sum-frequency generation (SFG). This technique enabled them to investigate the vibrational and electronic properties at the interfaces within operating OLEDs, offering new insights into their behavior under real-world conditions.

When voltage is applied to an OLED system, light is emitted via the recombination of charges at the organic interfaces. This alters the SFG output, allowing researchers to study how charge accumulates and what electronic structural changes occur at the interfaces under different operating conditions.

three different multilayer OLEDs with different types and combinations of organic layers were used. Electronic SFG (ESFG) spectroscopy was conducted on three OLED devices to examine spectral changes induced by the charge behavior and electronic structure at the interfaces. “We examined the differences in the electric field intensities inside the OLED devices based on the applied voltage dependence of the ESFG spectra. This clarifies the role of field strength differences that affect the ease of internal charge flow and the light emission characteristics for the first time,” explains Prof. Miyamae about the team’s study.

ESFG spectral bands corresponding to each organic layer were identified by comparing the absorption spectra and layer configurations among the three OLED devices. The researchers observed changes in spectral signal intensities when applying voltages to the OLED devices, which were related to the changes in the electric field and charge behavior inside the OLEDs.

Upon voltage application, the spectral signal intensity increased at the absorption band of the hole transport material (positive charge carriers inside the OLED), and signal intensity decreased at the absorption band of the light-emitting layer. This shows that internal charge flow across the organic layers within the OLEDs is different, leading to changes in spectra.

The team also applied square-wave pulse voltages on these devices to study how electric fields formed within these devices varied with time. They found that adding BAlq (a material used for electron transport in OLEDs) changes the position where the light is emitted in OLEDs. This shift in emission affects both the color and shape of the emitted light and how efficiently the device converts electricity into light.

“ESFG technique represents a novel, highly effective, nondestructive, and non-invasive spectroscopic approach for examining the electric field generation caused by injected charges in solid-state thin-film devices,” notes Prof. Miyamae about this innovative research work.

With this technique, material scientists can now design OLEDs with improved device lifetimes, energy efficiency, and cost reductions, eventually increasing the use of ultrathin organic devices in our everyday lives. “Moreover, this research can greatly shorten and rationalize materials development research, which now performs trial-and-error processes and long periods of degradation verification to assess device efficiency and lifetime,” adds Prof. Miyamae.

For more information: Journal of Materials Chemistry C

LIFT launches Advanced Metallic Production and Processing (AMPP)

LIFT, a Department of Defense-supported national advanced materials manufacturing innovation institute, has opened the Advanced Metallic Production and Processing (AMPP) Center in Detroit’s Corktown district. This facility will enhance the U.S. industrial base by accelerating the design, development, and deployment of novel metallic materials, addressing a critical gap in defense manufacturing. By producing metals across all alloy classes and processing them into high-quality feedstocks, AMPP will expedite materials development, particularly for additive manufacturing, ensuring manufacturers have access to essential materials for next-generation defense and commercial technologies.

Advanced materials are critical across industries. According to a recent report from Siemens Digital Industries Software, ”Innovation is key to the survival and growth of all companies. Product and materials inno­vation are closely linked, with product innovation relying on materials innovation by almost 70 percent. Therefore, it’s clear that developing new materials or innovative applica­tions of existing materials are of paramount importance for the manufacturing industry.”

This capability de-risks materials development investment for our domestic manufacturers and unlocks faster time to market and overall reduced cost of development.

A National Collaboration to Drive Innovation
The AMPP Center will serve as a hub for collaboration between LIFT’s nearly 400-member network and key industry stakeholders, including:
– Original Equipment Manufacturers
– Systems Manufacturers
– Materials Producers & Developers
– Application Developers & Part Manufacturers
– Academia & Startups

LIFT’s new Advanced Metallics Production and Processing Center exemplifies the mission of the Department of Defense’s Manufacturing Innovation Institutes: to close critical advanced manufacturing gaps in the U.S. defense industrial base and ensure we remain the strongest and most lethal force in the world,” said Keith DeVries, Director of Manufacturing Technology (ManTech) under the Office of the Under Secretary of Defense for Research and Engineering (OUSD(R&E)).

Advancing American Manufacturing & Security
Nigel Francis, CEO & Executive Director of LIFT, emphasized the importance of AMPP in maintaining America’s competitive edge:
“The pace of advanced manufacturing innovation is accelerating, and the development of novel materials is crucial to keeping the U.S. ahead of global competitors. With AMPP, we can now rapidly move new materials from concept to prototype, feasibility testing, and full-scale production—all within our borders. This capability is a game-changer for both our warfighters and manufacturers across industries.”

“A decade ago, we welcomed LIFT to Detroit to research and develop the next generation of advanced materials for commercial and defense uses,” said Mike Duggan, Mayor, City of Detroit. “This expansion into the production and processing of these materials demonstrates the success of their work and is a great example of how Detroit is very much a national center for innovation.”

Key Capabilities of the AMPP Center
– Low to Medium Volume Material Production
– Accelerated Material Delivery & Availability
– Toll Processing & Contract Manufacturing
– Domestic Production Through a Nonprofit Public-Private Partnership

Bringing Next-Generation Materials to Market and Scaling-Up Additive Manufacturing

As LIFT celebrates its 11th year as a national manufacturing innovation institute, it continues to expand its impact and technologies across the country, recently opening a facility in Puerto Rico and exploring new expansion opportunities.

For more information: LIFT

Exploring quantum materials for a new generation of technology

Today’s technology, from computer chips to camera image sensors, relies heavily on silicon semiconductors, which have been shrinking for decades. However, physical limitations will soon halt further advancements. Consequently, scientists and engineers are developing a new generation of technology based on quantum mechanics. Electrons in “quantum materials” exhibit unique behaviors, such as magnetism and superconductivity, which are crucial for future quantum technologies.

“Our piece of the puzzle is understanding how these materials function as a prerequisite for using them in engineering devices,” said Mark Dean, a physicist at the U.S. Department of Energy’s (DOE) Brookhaven National Laboratory and leader of the Dynamics and Control Group in Brookhaven’s Condensed Matter Physics and Materials Science Department.

Dean characterizes quantum materials using a technique called resonant inelastic X-ray scattering, or RIXS. RIXS is particularly suited for probing samples as thin as one atomic layer and material states that change very rapidly. And with recent technological developments, researchers expect this technique to enable studies that were unthinkable only five years ago.

This progress gave Dean and three colleagues — Matteo Mitrano, Steven Johnston, and Young-June Kim — the impetus to chart where the field is going as a whole. So, they summarized the technique’s state of the art and how they expect the field to progress in a Perspective paper.

For more information: Physical Review X

Image: The graphic illustrates an ultrabright X-ray striking a quantum material’s electrons (gray circles) and scattering off the sample. In this example, the X-ray’s energy change will provide insight into an electron property called spin, represented by the arrows. (Brad Baxley/Part to Whole LLC)

Scientists observe exotic quantum phase once thought impossible

Rice University researchers have directly observed a superradiant phase transition (SRPT), a quantum phenomenon predicted over 50 years ago, which could revolutionize quantum computing, communication, and sensing. This occurs when quantum particles fluctuate collectively without external triggers, forming a new state of matter. The discovery was made in a crystal of erbium, iron, and oxygen, cooled to minus 457°F and exposed to a magnetic field of up to 7 tesla.

“Originally, the SRPT was proposed as arising from interactions between quantum vacuum fluctuations — quantum light fields naturally existing even in completely empty space — and matter fluctuations,” said Dasom Kim, a Rice doctoral student in the Applied Physics Graduate Program who is a lead author on the study. “However, in our work, we realized this transition by coupling two distinct magnetic subsystems — the spin fluctuations of iron ions and of erbium ions within the crystal.”

Spin describes the magnetic poles of electrons or other particles and can be envisioned as a tiny arrow attached to each particle, constantly twirling and pointing in a given direction. When spins align, they create magnetic patterns across a material. When the pattern of spins ripples across the material like a wave, the resulting collective excitation is known as a magnon.

Until now, whether or not an SRPT could actually take place was subject to debate as it runs against a limitation — called “no-go theorem” in theoretical physics — arising in light-based systems. By staging an SRPT in a magnetic crystal based on the interactions between two spin subsystems, the researchers were able to get around this barrier, creating a magnonic version of the phenomenon. Specifically, the iron ions’ magnons play the role traditionally attributed to vacuum fluctuations, and the erbium ions’ spins represent matter fluctuations.

Using advanced spectroscopic techniques, the researchers observed unmistakable signatures of an SRPT, with the energy signal of one spin mode vanishing and another showing a clear shift or kink. These spectral fingerprints match exactly what theory predicts for entering the superradiant phase, giving the team high confidence that they had indeed coaxed the long-sought state into being.

“We established an ultrastrong coupling between these two spin systems and successfully observed a SRPT, overcoming previous experimental constraints,” Kim said.

Researchers are excited not just because a 50-year-old physics prediction has been confirmed but also because of what this could mean for quantum technology. Collective quantum states at the SRPT have unique properties that could be harnessed for next-generation quantum technologies.

“Near the quantum critical point of this transition, the system naturally stabilizes quantum-squeezed states — where quantum noise is drastically reduced — greatly enhancing measurement precision,” Kim said. “Overall, this insight could revolutionize quantum sensors and computing technologies, significantly advancing their fidelity, sensitivity and performance.”

Sohail Dasgupta, a graduate student at Rice working with Kaden Hazzard, associate professor of physics and astronomy, theoretically modeled the SRPT, building on a model developed by their collaborator and co-author Motoaki Bamba, a professor at Yokohama National University.

“Although the basic mathematical model was already laid out before by Motoaki, we needed to account for some of the specific magnetic properties of the material to obtain the precise results. When your theory matches the experimental data ⎯ which happens rather rarely ⎯ it is the best feeling for a scientist,” Dasgupta said.

Hazzard said the achievement shows that concepts from quantum optics can be translated into solid materials.

“This opens a new way to create and control phases of matter using ideas from cavity quantum electrodynamics,” Hazzard said.

Moreover, the crystal used in this study is one example of a broader class of materials, which means the research paves the way for exploring quantum phenomena in other materials with similarly interacting magnetic components.

For more information: Science Advances

Image: Dasom Kim (Photo by Jorge Vidal/Rice University)

SLAC fired the most intense submicron electron beam in history

Scientists at SLAC National Accelerator Laboratory have developed an ultrashort electron beam with five times the peak current of any previous beam, addressing a major challenge in particle accelerator and beam physics: generating high-power electron beams without compromising quality. This breakthrough opens new research possibilities in quantum chemistry, astrophysics, and materials science.

“Not only can we create such a powerful electron beam, but we’re also able to control the beam in ways that are customizable and on demand, which means we can probe a much wider range of physical and chemical phenomena than ever before,” said Claudio Emma, a staff scientist at the Department of Energy’s SLAC National Accelerator Laboratory, who is a researcher at SLAC’s Facility for Advanced Accelerator Experimental Tests (FACET-II) and a lead author on the new study.

One of the field’s longstanding goals has been to develop electron beams that are both extremely powerful and precisely controlled. Until now, increasing a beam’s power often meant degrading its quality, a tradeoff that has limited progress in many advanced experiments.

Traditionally, a microwave field is used to compress and focus the electron beam. The electrons within the field are staggered, so that those further back have more energy than those in the front. It’s sort of like runners staggered at the start of a track race, Emma explained. “We then send them around a bend, so the electrons in back catch up with electrons in front, and then at the end, you have a bunch of electrons together in a focused beam.”

The problem with this approach is that as they accelerate, electrons emit radiation and lose energy, so the quality of the beam deteriorates. That creates a tradeoff between beam energy and quality. “We can’t apply traditional methods to compress bunches of electrons at the submicron scale, while also preserving beam quality,” Emma said.

To solve this issue, SLAC researchers compressed billions of electrons into a length less than one micrometer using a laser-based shaping technique originally developed for X-ray free-electron lasers, such as SLAC’s Linac Coherent Light Source (LCLS). “The big advantage of using a laser is that we can apply an energy modulation that’s much more precise than what we can do with microwave fields,” Emma said.

But it’s not as simple as just shooting a few lasers down a tunnel. “We have a one-kilometer-long machine, and the laser interacts with the beam in the first 10 meters, so you have to get the shaping exactly right, then you have to transport the beam for another kilometer without losing this modulation, and you have to compress it,” Emma said. “So it wasn’t easy.”

After several months of testing and finessing their laser shaping technique, Emma and his team can now repeatedly produce high energy, femtosecond-duration, petawatt peak power electron beams that are about five times higher in current than what could previously be achieved.

This new beam will allow scientists to probe a whole series of natural phenomena, including testing hypotheses in quantum physics, materials science, and astrophysics.

In astrophysics, for example, this beam can be directed to a solid or gas target to create a filament similar to those seen in stars. “Scientists know that these filaments occur, but now we can test how they occur and evolve in the lab with a level of power we haven’t had before,” Emma said.

Fellow FACET-II researchers pounced on the more powerful beam and have already applied it to advancing plasma wakefield technology. Emma is particularly excited about the prospect of further compressing these beams to make attosecond light pulses, further enhancing LCLS’s current attosecond capabilities and driving even more pioneering science. “If you have the beam as a fast camera, then you also have a light pulse that’s very short, and now suddenly you have two complementary probes,” Emma explained. “That’s a unique capability and we can do a lot of things with that.”

Emma and his colleagues are excited about the prospects this new electron beam will bring. “We have a really exciting and interesting facility at FACET-II where people can come and do their experiments,” he said. “If you need an extreme beam, we have the tool for you, and let’s work together.”

For more information: Physical Review Letters

Image: At the heart of the new study is a laser heater undulator, a device that allows researchers to tightly control electron beams.

Twisting 2D materials creates artificial atoms that could advance quantum computers

Researchers at the University of Rochester have discovered that twisting two atom-thick flakes of special materials at high angles reveals unique optical properties, potentially useful for quantum computers and other quantum technologies. By precisely layering these nano-thin materials, they create excitons—artificial atoms that can function as quantum information bits, or qubits.

“If we had just a single layer of this material we’re using, these dark excitons wouldn’t interact with light,” says Nickolas Vamivakas, the Marie C. Wilson and Joseph C. Wilson Professor of Optical Physics. “By doing the big twist, it turns on artificial atoms within the material that we can control optically, but they are still protected from the environment.”

The work builds on the 2010 Nobel Prize–winning discovery that peeling carbon apart until it reaches a single layer of atoms creates a new two-dimensional (2D) material called graphene with special quantum characteristics.

Scientists have since explored how the optical and electrical properties of graphene and other 2D materials change when layered on top of one another and twisted at very small angles—called moiré superlattices. For example, when graphene is twisted at the “magic” angle of 1.1 degrees, it creates special patterns that produce properties such as superconductivity.

But scientists from Rochester’s Institute of Optics and Department of Physics and Astronomy took a different approach. They used molybdenum diselenide, a 2D material that is more fickle than graphene, and twisted it at much higher angles of up to 40 degrees. Still, the researchers found the twisted monolayers produced excitons that were able to retain information when activated by light.

“This was very surprising for us,” says Arnab Barman Ray, an optics Ph.D. candidate. “Molybdenum diselenide is notorious because other materials in the family of moiré materials show better information-retaining capacity. We think that if we use some of those other materials at these large angles, they will probably work even better.”

The team views this as an important early step toward new types of quantum devices.

“Down the line, we hope these artificial atoms can be used like memory or nodes in a quantum network, or put into optical cavities to create quantum materials,” says Vamivakas. “These could be the backbone for devices like the next generation of lasers or even tools to simulate quantum physics.”

For more information: Nano Letters

Developing 3D-printed soft material actuators that can mimic real muscles

Empa researchers have developed a 3D printing method to produce soft, elastic, yet powerful artificial muscles. These artificial muscles could one day be used in medicine, robotics, and other applications requiring movement at the touch of a button. While they hold potential for supporting people at work, aiding mobility, or replacing injured muscle tissue, creating artificial muscles that match the performance of real muscles remains a significant technical challenge.

In order to keep up with their biological counterparts, artificial muscles must not only be powerful, but also elastic and soft. At their core, artificial muscles are so-called actuators: Components that convert electrical impulses into movement. Actuators are used wherever something moves at the push of a button, whether at home, in a car engine or in highly developed industrial plants. However, these hard mechanical components do not have much in common with muscles just yet.

The dielectric elastic actuators (DEA) consist of two different silicone-based materials: a conductive electrode material and a non-conductive dielectric. These materials interlock in layers. “It’s a bit like interlacing your fingers,” explains Empa researcher Patrick Danner. If an electrical voltage is applied to the electrodes, the actuator contracts like a muscle. When the voltage is switched off, it relaxes to its original position.

3D printing such a structure is not trivial, Danner knows. Despite their very different electrical properties, the two soft materials should behave very similarly during the printing process. They should not mix but must still hold together in the finished actuator.

The printed “muscles” must be as soft as possible so that an electrical stimulus can cause the required deformation. Added to this are the requirements that all 3D printable materials must fulfill: They must liquefy under pressure so that they can be extruded out of the printer nozzle. Immediately thereafter, however, they should be viscous enough to retain the printed shape.

“These properties are often in direct contradiction,” says Danner. “If you optimize one of them, three others change … usually for the worse.”

In collaboration with researchers from ETH Zurich, Danner and Dorina Opris, who leads the research group Functional Polymeric Materials, have succeeded in reconciling many of these contradictory properties. Two special inks, developed at Empa, are printed into functioning soft actuators using a nozzle developed by ETH researchers Tazio Pleij and Jan Vermant.

The collaboration is part of the large-scale project Manufhaptics, which is part of the ETH Domain’s strategic area Advanced Manufacturing. The aim of the project is to develop a glove that makes virtual worlds tangible. The artificial muscles are designed to simulate the gripping of objects through resistance.

However, there are far more potential applications for soft actuators. They are light, noiseless and, thanks to the new 3D printing process, can be shaped as required. They could replace conventional actuators in cars, machinery and robotics. If they are developed even further, they could also be used for medical applications.

Opris and Danner are already working on it. Their new process can be used to print not only complex shapes, but also long elastic fibers. “If we manage to make them just a little thinner, we can get pretty close to how real muscle fibers work,” says Opris. The researcher believes that in the future it may be possible to print an entire heart from these fibers. However, there is still a lot to do before such a dream becomes a reality.

For more information: Advanced Materials Technologies

Image: Complexity on a small scale: A 3D-printed soft actuator or “artificial muscle.”

Penn State to establish new advanced semiconductor lab

Penn State researchers are set to enhance their semiconductor technology R&D capabilities with $4.3 million in funding and support from MMEC, a consortium focused on microelectronics. This funding, part of the Department of Defense’s Microelectronics Commons initiative under the CHIPS Act, will help establish an advanced lab for semiconductor thin films and device research at the Materials Research Institute in the Millennium Science Complex. MMEC, founded by Battelle, unites industry, academia, and government to drive innovation in microelectronics for commercial and defense applications, strengthening the U.S. supply chain.

“We were very fortunate to be included in the original MMEC proposal,” said Joan Redwing, lead investigator on the infrastructure project and distinguished professor of materials science and engineering and director of MRI’s Two-Dimensional Crystal Consortium, a U.S. National Science Foundation Materials Innovation Platform and national user facility. “The proposal included infrastructure investment for training and workforce development. The funding will allow MRI to build capacity for next-generation semiconductor thin films and devices, which includes new equipment that will allow us to scale up fabrication and create prototype devices.”

At the heart of the new facility, made possible by the funding, will be a metal-organic chemical vapor deposition (MOCVD) tool, manufactured by AIXTRON SE, a multinational technology company. The MOCVD tool works by heating a chamber in a highly controlled manner where special chemical gases, containing the elements needed for the material, are introduced. These gases react and break down on a hot surface, such as a semiconductor wafer, depositing a thin, even layer of material. This precise layering allows for high-quality materials used in advanced technologies like semiconductors.

This instrument will enable the deposition of semiconductor thin films on multiple wafers at a time at sizes up to four-inch diameter. The tool is unique in its ability to grow both wide bandgap semiconductors such as gallium nitride — used in power electronics — and two-dimensional (2D) materials — an emerging ultra-thin semiconductor for logic and brain-inspired computing. Gallium nitride and 2D materials have applications in high-performance power electronics and energy-efficient computing, which are critical technologies for electric vehicles and artificial intelligence, among other applications.

“This tool will allow students and early career researchers to gain hands-on experience with state-of-the-art thin film deposition equipment used by industry for compound semiconductor thin film manufacturing,” Redwing explained. “It will also provide new capabilities for scaling up thin film materials for device research, particularly for advanced semiconductors including wide bandgap and 2D materials.”

In addition to the MOCVD tool, the lab will house several other specialized instruments. One is a Jupiter XR atomic force microscope from Oxford Instruments Asylum Research for fast scanning and full-wafer mapping, which will help enhance quality control and characterization of thin film materials. The other is an evaporator for deposition of specialized contact metal stacks for devices fabricated using 2D materials. This tool will support research by Suzanne Mohney, professor of materials science and engineering, and Saptarshi Das, professor of engineering science and mechanics, in addition to other faculty.

The equipment will be available for use by researchers inside and outside Penn State as a user facility with shared process know-how. The lab’s capabilities will provide opportunities for a range of Penn State researchers, including faculty working on power electronics and 2D-device development.

“This new lab connects us more closely with MMEC and provides a unique opportunity to support training and workforce development as well as collaborative research with universities and industry partners across the consortium,” Redwing said.

For more information: Penn State University

Image: The new lab will enable work developing advanced semiconductor wafers, shown here.

3D-printed knee implants improves quality and reliability

Researchers at Naton Biotechnology have advanced 3D-printed medical implants by developing the world’s first laser 3D-printed total knee implant, which has been approved by China’s National Medical Products Administration. Their study focused on enhancing the strength and consistency of cobalt-chromium-molybdenum (CoCrMo) alloy implants using laser powder bed fusion (LPBF). By optimizing heat treatment, they corrected material inconsistencies, resulting in stronger, more reliable, and safer implants for patients.

This research provides key insights into how 3D printing affects metal implants and lays the foundation for better quality control in orthopedic manufacturing, helping to advance the future of customized medical implants.

This research was led by Professor Changhui Song from South China University of Technology and Professor Jia-Kuo Yu from Beijing Tsinghua Changgung Hospital as co-corresponding authors. The study was conducted in collaboration with Senior Engineer Renyao Li from Naton Biotechnology (Beijing) Co., Ltd and other members of the team.

The layer-by-layer manufacturing process of CoCrMo, a widely used implant material, occurs at extremely high cooling rates (~10⁵–10⁶ K/s). This rapid solidification often leads to anisotropy, meaning the material’s properties vary depending on the direction of force. The main causes include columnar grain structures, porosity, and residual stress, all of which are inherent to additive manufacturing.

While extensive research has been conducted on LPBF-fabricated CoCrMo alloys, most studies have only examined their performance in a single direction, overlooking how anisotropy affects overall durability. However, implants inside the human body must withstand forces from multiple directions. Then, if the material’s strength is inconsistent, weak spots can develop, increasing the risk of breakage or failure.

In mechanical tests, CoCrMo samples stretched significantly more in one direction (19.1% elongation) than in another (9.3% elongation)—a disparity of over 100%. This inconsistency makes the material unreliable for long-term medical use, as implants must perform uniformly and safely under everyday stresses.

The team found that a two-step heat treatment process significantly improved the uniformity of the metal’s structure and strength. The process included:

  • Solution Treatment – Heating the material to 1150°C, holding it for an hour, and then rapidly cooling it in water. This helped restructure the uneven metal grains.
  • Annealing – Reheating the material to 450°C for 30 minutes and then cooling it again. This step refined the grain structure and further balanced the material’s properties.

As a result, the metal’s strength and flexibility became nearly identical in all directions. The ultimate tensile strength reached 906.1 MPa and 879.2 MPa, while elongation values balanced at 20.2% and 17.9%, making the material stronger and more reliable for medical use.

With this breakthrough, scientists are now looking at surface treatments to further enhance the wear resistance and biocompatibility of implants. Methods like shot peening (where tiny metal beads are blasted onto the surface) and ultrasonic peening could improve the fatigue resistance of implants, helping them last longer under daily stress. These next-generation treatments could make 3D-printed joint implants even more durable and widely used in clinical settings.

This research offers new insights into how to improve 3D-printed metal implants, making them safer and more durable for patients. By addressing uneven strength and material quality, this breakthrough lays the foundation for better orthopedic implants, particularly for joint replacements.

For more information: IOP Science

Image: Diagrams illustrating how columnar grains develop during the LPBF, contributing to microstructural anisotropy; (b) Stress-strain curves showing significant improvements in the mechanical anisotropy after heat treatment; (c) Diagrams showing the recrystallization process, where equiaxed grains form to eliminate directional effects and enhance uniformity; (d) TEM images revealing the nanoscale interactions between martensite laths in the solution-annealed state; (e) Visual representation of the synergistic effects between annealing twins and martensite laths.

Rare-earth-free solar cells could lower costs and boost accessibility

Researchers at the University of Sheffield, in collaboration with Power Roll Ltd., have developed a cost-effective, flexible solar cell using a perovskite semiconductor. This innovative design, which consolidates electrical contacts onto the back of the cell, simplifies production and enhances efficiency, making solar power more accessible, especially in regions where conventional panels are impractical.

Continue reading

US DOE earmarks $179M for microelectronics science research centers

The U.S. Department of Energy (DOE) announced $179 million in funding over four years for three Microelectronics Science Research Centers (MSRCs), authorized by the Micro Act in the CHIPS and Science Act of 2022. These centers will conduct basic research in microelectronics materials, device and system design, and manufacturing science to transform future microelectronics technologies. The MSRCs, formed as networks of 16 projects led by 10 national laboratories, complement activities under the CHIPS and Science Act at the Department of Commerce, the Department of Defense, and other agencies.

The Microelectronics Energy Efficiency Research Center for Advanced Technologies (MEERCAT) will advance integrated innovations across materials, devices, information-carrying modalities, and systems’ architectures. Focusing on intelligent sensing, data bandwidth, multiplexing, and advanced computing, the center will explore transformative solutions that seamlessly bridge sensing, edge processing, artificial intelligence, and high-performance computing.

The Co-design and Heterogeneous Integration in Microelectronics for Extreme Environments (CHIME) Center will develop extreme environment electronics through heterogeneous integration and a seamless fusion of diverse materials, processes, and technologies to enable next-generation systems. The center will create robust, high-performance solutions capable of excelling in the most challenging conditions, including extreme thermal and radiation environments.

The Extreme Lithography & Materials Innovation Center (ELMIC) aims to advance the fundamental science driving the integration of new materials and processes into future microelectronic systems, focusing on key areas such as plasma-based nanofabrication, extreme UV photon sources, 2D-material systems, and extreme-scale memory. The center’s scientific investigations will be informed strongly by a systems-to-physics motivation aimed at long term impact.

For more information: U.S. Department of Energy

Image: (From left) Lawrence Livermore National Laboratory researchers Drew Willard, Brendan Reagan, and Issa Tamer work on the Big Aperture Thulium laser system, one of the projects designated under the Extreme Lithography & Materials Innovation Center.

Behold the world’s thinnest pasta

The world record for thinnest pasta has just been shattered.

A team of researchers has created starchy nanofibers from white flour, with an average thickness of about 370 nanometers, or roughly two-hundredths the thickness of a human hair. These nanofibers, produced by mixing flour with formic acid to uncoil the starch molecules, are being explored for use in biodegradable bandages rather than as food.

“Normally, if you want to cook starch, then you use water and heat to break up the tight packing of starch,” notes Adam Clancy. “We do that chemically with formic acid. So we effectively pickle it instead of cooking it.”

Clancy is a chemist at University College London in England. He was part of a team that warmed this dough to give it the right consistency. To stretch the dough into tiny noodles, they used a technique called electrospinning.

In this process, an electrical charge pulled the dough through a needle and onto a plate several centimeters away. The starch molecules tangled with each other as they left the needle, forming a jet. As the jet flew through the air, the formic acid evaporated. This left behind a thin fiber. After about 30 minutes, the fiber formed a thin mat on the plate.

This isn’t the first time someone has made a mat of starchy nanofibers. Such mats typically have tiny holes called pores. These pores are large enough to let water molecules through, but too small for bacteria to enter. That makes them good options for bandages and wound dressings.

But past research has electrospun mats using pure starch, as opposed to a starch-containing flour. The process of extracting pure starch from plant matter takes a lot of energy and water. The new study shows such costly extraction isn’t always needed.

Since the fibers are made of dried flour, they count as pasta, the authors say. That makes them the thinnest pasta on record.

Each noodle is roughly a thousandth the width of su filindeu. That’s a type of pasta about half the width of angel hair noodles, which are about 1 millimeter (0.04 inch) thick when cooked.

So is Clancy’s nanopasta edible? “I certainly hope so,” he says.

For more information: Nanoscale Advances

Image: These nanofibers (seen under a scanning electron microscope) can be called pasta because they are made from dried flour. That makes them the thinnest noodles ever.

Scientists use quantum computers to simulate elusive particles

Researchers from Harvard University, MIT, and QuEra Computing Inc. have developed a new digital quantum simulation architecture based on reconfigurable atom arrays to simulate fermionic systems on quantum computers. This approach aims to advance materials science, chemistry, and high-energy physics by improving simulations of complex quantum materials and exotic phases of matter, which are challenging to study with classical computational methods. Fermions, such as electrons, protons, and neutrons, obey the Pauli exclusion principle, making their simulation difficult due to non-local interactions where changes in one part of the system can affect distant parts.

The research team used a quantum computing method that maps fermionic behavior onto qubits by leveraging a model known as Kitaev’s honeycomb lattice. This mathematical framework describes how particles interact in a two-dimensional honeycomb-shaped grid. The Kitaev model is particularly useful because it allows researchers to study exotic phases of matter, including spin liquids — materials where magnetic moments remain disordered even at very low temperatures. The approach encodes fermionic statistics using long-range entangled states to allow for more efficient quantum simulations of strongly interacting systems.

The study confirms the existence of a non-Abelian spin liquid phase by measuring an odd Chern number, a key mathematical indicator of topological order. Non-Abelian states are special because the way particles interact depends on the order in which they are exchanged. Unlike conventional particles, which either return to their original state or flip when swapped, non-Abelian particles retain a memory of the sequence of swaps, making them useful for robust quantum computing.

Although it’s not exact, to grasp some idea of the concept, you could think of non-Abelian states like a combination lock — if you enter the numbers in a different order, it will give you a different result. On the other hand, conventional particles are like light switches, which only have two states regardless of how many times you flip them.

The odd Chern number serves as a topological signature that identifies different quantum phases, much like how a fingerprint identifies an individual.

The work demonstrates how quantum simulations can efficiently capture the essential physics of strongly correlated fermions, according the researchers.

By encoding fermionic interactions in a topologically ordered state, the system enables precise control over the quantum evolution of fermions. Topological states are quantum states of matter that depend on their overall structure rather than specific local details. These states are particularly important because they provide robust ways to store and manipulate quantum information, making them valuable for fault-tolerant quantum computing.

Quantum simulations of fermionic systems are particularly challenging due to the non-local nature of fermion interactions. Traditional quantum computing methods require complex encodings that can introduce significant computational overhead. The researchers overcame these challenges using measurement-based state preparation, tunable Floquet circuits, and error detection techniques. Floquet engineering, which is a method that uses periodic driving to control quantum interactions, played a crucial role in simulating fermionic behavior in a controllable manner.

In the experiment, the team used a reconfigurable array of 104 atomic qubits on a neutral-atom quantum computing platform to represent the honeycomb lattice. They employed Floquet engineering to simulate fermionic behavior. Through this method, they successfully prepared and verified various low-energy states, including those corresponding to topological spin liquids. These spin liquids are phases of matter that, unlike conventional magnets, do not exhibit long-range magnetic order but instead host highly entangled quantum states with exotic properties such as fractionalized excitations.

One of the study’s major findings is the ability to simulate strong interactions within a fermionic system, particularly in the context of the Fermi-Hubbard model. This model is widely used in condensed matter physics to describe electron interactions in materials and has implications for understanding high-temperature superconductivity. By engineering interactions among fermions on a square lattice, the researchers were able to explore dynamics relevant to real-world materials, potentially providing insights into novel electronic properties that could be harnessed in future technologies.

Despite these advances, the study acknowledges several limitations that will likely be the focus of future work, as we’ll see later. Quantum errors remain a significant challenge, limiting the depth of circuits that can be executed before decoherence affects results. To mitigate these errors, the researchers implemented built-in error detection methods and post-selection techniques, improving the accuracy of their results. However, fully error-corrected simulations will require further advancements in quantum hardware and fault-tolerant encoding schemes. Scaling up these simulations to more complex and larger systems will also require improving quantum hardware stability and reducing noise.

Future directions for this research include expanding the scale of quantum simulations and integrating more sophisticated error correction techniques. The researchers also suggest that their approach could be applied to other complex quantum systems, including lattice gauge theories and quantum gravity models. Lattice gauge theories are mathematical frameworks used to describe fundamental interactions in particle physics, and their simulation could provide new insights into quantum chromodynamics, the theory governing the strong force that binds atomic nuclei.

Simulating aspects of quantum gravity could shed light on how quantum mechanics and general relativity interact at small scales, an area that remains largely unexplored due to the limitations of classical computation.

For more information: arXiv

Designing nano-architected materials using ML and 3D printing

Researchers at the University of Toronto’s Faculty of Applied Science & Engineering have used machine learning and 3D printing to create nano-architected materials that combine the strength of carbon steel with the lightness of Styrofoam. In a new paper, Professor Tobin Filleter’s team describes these nanomaterials, which offer exceptional strength, light weight, and customizability, potentially benefiting industries from automotive to aerospace.

“Nano-architected materials combine high-performance shapes, like making a bridge out of triangles, at nanoscale sizes, which takes advantage of the ‘smaller is stronger’ effect, to achieve some of the highest strength-to-weight and stiffness-to-weight ratios, of any material,” said Peter Serles, the first author of the new paper. “However, the standard lattice shapes and geometries used tend to have sharp intersections and corners, which leads to the problem of stress concentrations. This results in early local failure and breakage of the materials, limiting their overall potential. “As I thought about this challenge, I realized that it is a perfect problem for machine learning to tackle.”

Nano-architected materials are made of tiny building blocks or repeating units measuring a few hundred nanometres in size – it would take more than 100 of them patterned in a row to reach the thickness of a human hair. These building blocks, which in this case are composed of carbon, are arranged in complex 3D structures called nanolattices.

To design their improved materials, Serles and Filleter worked with Professor Seunghwa Ryu and PhD student Jinwook Yeo at the Korea Advanced Institute of Science & Technology (KAIST) in Daejeon, South Korea. This partnership was initiated through the University of Toronto’s International Doctoral Clusters program, which supports doctoral training through research engagement with international collaborators.

The KAIST team employed the multi-objective Bayesian optimization machine learning algorithm. This algorithm learned from simulated geometries to predict the best possible geometries for enhancing stress distribution and improving the strength-to-weight ratio of nano-architected designs.

Serles then used a two-photon polymerization 3D printer housed in the Centre for Research and Application in Fluidic Technologies (CRAFT) to create prototypes for experimental validation. This technology enables 3D printing at the micro and nanoscale – creating optimized carbon nanolattices.

These optimized nanolattices more than doubled the strength of existing designs – withstanding stress of 2.03 megapascals for every cubic meter per kilogram of its density, which is about five times higher than titanium.

“This is the first time machine learning has been applied to optimize nano-architected materials, and we were shocked by the improvements,” said Serles. “It didn’t just replicate successful geometries from the training data; it learned from what changes to the shapes worked and what didn’t, enabling it to predict entirely new lattice geometries. Machine learning is normally very data-intensive, and it’s difficult to generate a lot of data when you’re using high-quality data from finite element analysis. But the multi-objective Bayesian optimization algorithm only needed 400 data points, whereas other algorithms might need 20,000 or more. So, we were able to work with a much smaller but an extremely high-quality data set.”

“We hope that these new material designs will eventually lead to ultra-lightweight components in aerospace applications, such as planes, helicopters, and spacecraft that can reduce fuel demands during flight while maintaining safety and performance,” said Filleter.

“This can ultimately help reduce the high carbon footprint of flying. For example, if you were to replace components made of titanium on a plane with this material, you would be looking at fuel savings of 80 liters per year for every kilogram of material you replace,” said Serles.

“Our next steps will focus on further improving the scale-up of these material designs to enable cost-effective macroscale components,” said Filleter. “In addition, we will continue to explore new designs that push the material architectures to even lower density while maintaining high strength and stiffness.”

For more information: Advanced Materials

Sandia partners with national labs to develop energy-efficient AI and computing tech

To address future energy needs, the Department of Energy Office of Science has announced the creation of three new Microelectronics Science Research Centers. One of these, the Microelectronics Energy Efficiency Research Center for Advanced Technologies (MEERCAT), will focus on energy efficiency by exploring solutions that integrate sensing, edge processing, artificial intelligence, and high-performance computing. Sandia National Laboratories will be a founding member of MEERCAT and will lead one of its eight energy efficiency-related research projects.

The other two centers will work on resilience in extreme environments, including high-radiation, cryogenic and high magnetic field environments.

“Our center will provide industry with new, higher performance options for energy-efficient computing,” said Nelson, the principal investigator for the Sandia-led project.

Sandia is also partnering on two projects led by other laboratories: one on energy efficiency with Lawrence Berkeley National Laboratory and another on extreme environments with Los Alamos National Laboratory.

AI is a major factor in rising energy demand because it uses more energy than conventional computer algorithms and has seen a surge in popularity within homes and workplaces. Along with the growth of other energy-intensive technologies like quantum computing and advanced sensors, this has created an urgent need for more efficient technologies.

The three new research centers will provide a total of $179 million for 16 multidisciplinary, fundamental research projects lasting up to four years. They are funded through DOE’s Office of Science and authorized by the Micro Act, passed in the CHIPS and Science Act of 2022. This legislation has invested billions of dollars through multiple agencies to help companies build new plants for advanced semiconductors in the U.S. It also funds fundamental research to advance the technologies these future factories will produce.

“We are working with companies to understand their problems and pulling experts together from across the DOE to solve these problems quickly,” Nelson said.

When the Energy Department announced its plan to form Microelectronics Science Research Centers in May 2024, Nelson reached out to a familiar team.

Two years earlier, a group of directors and experts from DOE’s five scientific user facilities, the Nanoscale Science Research Centers, had started holding regular, collaborative discussions.

“We met every two weeks for two years,” Nelson said. “We discussed our collective resources and how we can work together to achieve national priorities.”

Nelson is the director of one of these Office of Science user facilities, the Center for Integrated Nanotechnologies, which is jointly operated by Sandia and Los Alamos national laboratories. The other four user facilities: the Center for Nanoscale Materials, the Center for Functional Nanomaterials, The Molecular Foundry and the Center for Nanophase Materials Sciences are spread across the country, each co-located at a national lab.

The team agreed that by working together they could advance new materials to make computing more powerful and energy-efficient.

Researchers had already found that materials like molybdenum disulfide, gallium arsenide and even diamond may be better than silicon for certain aspects of computing. In theory, computer chips made from one of these alternative materials might be far more energy-efficient and could solve the looming energy crisis.

“They’re very promising,” Nelson said.

But the task of taking any of these materials, perfecting them in a lab, learning how to mass produce them and then building a factory to make chips from them while competing against an established silicon industry and supply chain, the team agreed, felt daunting at best.

Taking a different route, the group of lab leads and other collaborators proposed a project entitled “Nano-Scale Research Center for Heterogeneous Integration Platforms.” This project would aim to leverage the existing infrastructure and expertise of the DOE user facilities and partnering institutions and develop ways to insert new materials into standard silicon fabrication processes.

Now greenlit with DOE’s recent announcement, the project will bring together resources from all five Nanoscale Science Research Centers. It will also include researchers from Fermi National Accelerator Laboratory, the Massachusetts Institute of Technology and MIT Lincoln Laboratory.

They will build on previous research in what scientists call heterogeneous integration. This means using many kinds of materials to make computer chips, all monolithically integrated into a silicon backbone. The tricky part is to ensure electrons and information flow seamlessly between different materials.

Sandia and its collaborators are aiming for breakthroughs that could help industry create much more energy-efficient computer chips.

“By collaborating across multiple national laboratories and universities, our goal is really to accelerate the innovation discovery process and make a positive impact on economic and national security,” Nelson said.

For more information: Sandia National Laboratories

Image: The Center for Integrated Nanotechnologies, pictured here, is one of five Department of Energy Nanoscale Science Research Centers teaming up to help make computer chips more energy-efficient.

Researchers develop revolutionary Diamond fabrication technology

A research team led by Professors Zhiqin Chu and Yuan Lin at the University of Hong Kong, in collaboration with Professors Kwai Hei Li and Qi Wang, has developed a groundbreaking method for producing ultrathin and ultra-flexible diamond membranes. These membranes are compatible with current semiconductor manufacturing processes, allowing their integration into various applications, including electronic, photonic, mechanical, acoustic, and quantum devices.

The team’s innovative edge-exposed exfoliation method allows for the rapid, scalable production of free-standing diamond membranes. This technique surpasses traditional methods, which are typically expensive, time-consuming, and limited in size. Notably, the new process can produce a two-inch diamond wafer in just 10 seconds, setting a new benchmark for efficiency and scalability in the field.

These ultra-flat diamond surfaces, essential for high-precision micromanufacturing, along with the flexibility of the membranes, open up new possibilities for next-generation flexible and wearable electronic and photonic devices. The research team envisions significant industrial applications in electronics, photonics, mechanics, thermics, acoustics, and quantum technologies.

“We hope to promote the usage of the high-figure-of-merit diamond membrane in various fields, and to commercialize this cutting-edge technology and deliver premium diamond membranes, setting a new standard in the semiconductor industry. We are eager to collaborate with academic and industry partners to bring this revolutionary product to market and accelerate the arrival of the diamond era,” concluded Professor Chu.

Diamonds, renowned globally as valuable gemstones, possess exceptional versatility in various scientific and engineering applications. They are the hardest natural material, boasting unparalleled thermal conductivity at room temperature, extremely high carrier mobility, dielectric breakdown strength, an ultrawide bandgap, and optical transparency spanning from the infrared to the deep-ultraviolet spectrum. These remarkable properties make diamonds ideal for fabricating advanced high-power, high-frequency electronic devices, photonic devices, and heat spreaders to cool high-power-density electronic components, such as those in processors, semiconductor lasers, and electric vehicles. However, the inert nature and rigid crystal structure of diamonds pose significant challenges in fabrication and mass production, particularly for ultrathin and freestanding diamond membranes, thereby restricting their widespread usage.

For more information: Nature

Google’s quantum error correction has some competition

Google Quantum AI’s significant advance in quantum error correction using a surface code approach faces competition from a rival method that proponents claim offers greater efficiency and scalability. Researchers are divided on which approach will shape the future of practical quantum computing. Quantum computers, promising solutions to complex problems in materials science, chemistry, and logistics, are extremely sensitive and prone to errors, which increase as the machines scale up, making error correction crucial for practical use.

Researchers at Google Quantum AI recently demonstrated that their quantum processor, Willow, could mitigate this issue using the surface code, a mathematical framework that groups physical qubits into “logical qubits.” This grouping protects calculations from errors without negatively impacting performance.

The Google Quantum AI team members recently made headlines when they reported that they were able to scale from a 3×3 grid to 5×5 and then to 7×7 grids of physical qubits reduced errors by a factor of two each time.

The method they used — called a surface code — has long been the dominant strategy for quantum error correction. It arranges qubits in interwoven grids, with data qubits performing calculations and ancillary qubits monitoring for errors. While effective, it requires a significant number of qubits to operate, which has limited its utility, according to New Scientist.

In 2023, IBM introduced a rival method called QLDPC (quantum low-density parity-check) code. Unlike the surface code, QLDPC connects each qubit to six others, allowing them to monitor each other’s errors. According to IBM researchers, this method could achieve the same error-correction capabilities as the surface code but with far fewer qubits. For example, on paper, where the surface code might require 4,000 qubits, QLDPC could deliver equivalent performance with just 288 qubits.

“With QLDPC, that lower qubit overhead is hard to compete with,” said Joe Fitzsimons of Horizon Quantum, a quantum computing startup.

IBM has tailored its quantum chips to support the connectivity demands of QLDPC. While adding these connections poses engineering challenges, IBM has reported that the changes do not compromise the reliability of its chips.

Oliver Dial, an IBM researcher, emphasized the importance of tailoring codes to the capabilities of specific hardware during a presentation at the Q2B conference in December.

The competition between the surface code and the theoretical QLDPC highlights a broader challenge in quantum computing: the interplay between hardware and software. Superconducting qubits, like those used by Google and IBM, are limited in how they can be connected, making some error-correction methods more practical than others.

However, alternative technologies, such as qubits made from ultracold atoms, could provide greater flexibility.

“Maybe someone somewhere is working on a type of surface code that is really great, but right now there is competition [to the surface code],” said Yuval Boger of QuEra Computing, a U.S.-based quantum startup.

The QuEra team previously worked with ultracold-atom qubits to achieve one of the largest groups of logical qubits, exploring various codes to optimize their usefulness.

Despite the excitement around QLDPC, the surface code remains a strong contender, Google’s team countered. Its theoretical framework is well understood, having been studied for more than two decades. It also offers a balance between performance and hardware requirements, making it particularly suitable for the superconducting qubits used in Google’s Willow processor.

“The surface code is well understood, with a well-studied theoretical framework. It offers a balance between performance and required qubit connectivity,” said Sergio Boixo of Google Quantum AI.

Google, however, is not resting on its laurels. Boixo confirmed that the team is exploring alternative error-correction codes alongside the surface code.

For more information: Nature

Scientists discover a way to shrink quantum computer components by 1,000X

Researchers have discovered a method to make quantum computing more compact, potentially shrinking essential components by 1,000 times and requiring less equipment. Current quantum computers rely on entangled photons produced by shining a laser on millimeter-thick crystals, but this setup is too large for integration into a computer chip.

Scientists at Nanyang Technological University, Singapore (NTU Singapore) have addressed this issue by producing entangled photon pairs using much thinner materials, just 1.2 micrometers thick, without needing additional optical gear to maintain the link, thereby simplifying the overall setup.

“Our novel method to create entangled photon pairs paves the way for making quantum optical entanglement sources much smaller, which will be critical for applications in quantum information and photonic quantum computing,” said NTU’s Professor Gao Weibo who led the researchers.

He added that the method could scale down the size of devices for quantum applications because many of these devices currently need large and bulky optical equipment, which are cumbersome to align, before they can work.

Quantum computers are expected to revolutionize the approach to many challenges, from helping us better understand climate change to finding new drugs faster by completing complex computations and quickly finding patterns in large data sets. For instance, calculations that would take supercomputers today millions of years to resolve could be done within minutes by quantum computers.

This is expected to happen because quantum computers perform many computations simultaneously instead of doing them one at a time like standard computers.

Quantum computers can do so as they perform calculations using tiny switches called quantum bits, or qubits, that can be in both the on and off position simultaneously. It is akin to flipping a coin in the air, with the spinning coin in a state between heads and tails. In contrast, standard computers use switches that can be on or off at any time, but not both.

Photons can be used as qubits for quantum computers to perform faster calculations as they can have on and off states at the same time. But being in two states simultaneously only happens if the photons are produced in a pair, with one photon linked, or entangled, to the other. An important condition for entanglement is that the paired photons need to vibrate in sync.

One advantage of using photons as qubits is that they can be produced and entangled at room temperature. Relying on photons can thus be easier, cheaper, and more practical than using other particles like electrons that need ultra-low temperatures close to the coldness of outer space before they can be used for quantum computing.

Researchers have been trying to find thinner materials to produce linked pairs of photons so that they can be worked into computer chips. However, one challenge is that when materials get thinner, they produce photons at a much lower rate, which is impractical for computing.

Recent advances showed that a promising new crystalline material called niobium oxide dichloride, which has unique optical and electronic properties, can produce pairs of photons efficiently despite its thinness. But these photon pairs are useless for quantum computers because they are not entangled when produced.

A solution was found by NTU scientists led by Professor Gao, from the University’s School of Electrical & Electronic Engineering and School of Physical & Mathematical Sciences, in collaboration with Professor Liu Zheng from the School of Materials Science & Engineering.

Professor Gao’s solution was inspired by an established method to create entangled pairs of photons with thicker and bulkier crystalline materials, which was published in 1999. It involves stacking two flakes of thick crystals together and positioning the crystalline grains of each flake perpendicularly to each other.

However, the vibrations of photons produced in a pair can still be out of sync due to how they travel within the thick crystals after they are created. Additional optical equipment is therefore needed to synchronize the photon pairs to maintain the link between the light particles.

Professor Gao theorized that a similar two-crystal set-up could be used with two thin crystal flakes of niobium oxide dichloride, with a combined thickness of 1.2 micrometers, to produce the linked photons without requiring extra optical instruments.

He expected this to happen because the flakes used are much thinner than the bulkier crystals from earlier studies. As a result, the pairs of photons produced travel a smaller distance within the niobium oxide dichloride flakes, so the light particles remain in sync with each other. Experiments by the NTU Singapore team proved that his hunch was correct.

Professor Sun Zhipei from Finland’s Aalto University, who specializes in photonics and was not involved in NTU’s research, said that entangled photons are like synchronized clocks that show the same time no matter how far apart they are and can thus enable instant communication.

He added that the NTU team’s method for generating quantum entangled photons “is a major advancement, potentially enabling the miniaturization and integration of quantum technologies.”

“This development has potential in advancing quantum computing and secure communication, as it allows for more compact, scalable, and efficient quantum systems,” said Professor Sun, a co-principal investigator at the Research Council of Finland’s Center of Excellence in Quantum Technology.

The NTU team plans to further optimize the design of their setup to generate even more linked pairs of photons than are currently possible.

Some ideas include exploring whether introducing tiny patterns and grooves on the surface of niobium oxide dichloride flakes can increase the number of photon pairs produced. Another one will examine whether stacking the niobium oxide dichloride flakes with other materials can boost photon production.

For more information: Nature Photonics

Image: PhD student Leevi Kallioniemi from NTU Singapore’s School of Physical & Mathematical Sciences with a blue laser set-up for generating entangled photon pairs. Credit: NTU Singapore

Nanostructures pave the way for advanced robotics

Researchers at the University of Sydney Nano Institute have made a significant advance in molecular robotics by developing custom-designed and programmable nanostructures using DNA origami, an innovative method that leverages the natural folding power of DNA to create new and useful biological structures. This approach has potential applications in targeted drug delivery systems, responsive materials, and energy-efficient optical signal processing. As a proof-of-concept, the researchers created over 50 nanoscale objects, including a “nano-dinosaur,” a “dancing robot,” and a mini-Australia that is 150 nanometers wide, a thousand times narrower than a human hair.

The research, led by first author Dr. Minh Tri Luu and research team leader Dr. Shelley Wickham, focuses on the creation of modular DNA origami “voxels” that can be assembled into complex three-dimensional structures. (Where a pixel is two-dimensional, a voxel is realized in 3D.)

These programmable nanostructures can be tailored for specific functions, allowing for rapid prototyping of diverse configurations. This flexibility is crucial for developing nanoscale robotic systems that can perform tasks in synthetic biology, nanomedicine and materials science.

Dr. Wickham, who holds a joint position with the Schools of Chemistry and Physics in the Faculty of Science, said, “The results are a bit like using Meccano, the children’s engineering toy, or building a chain-like cat’s cradle. But instead of macroscale metal or string, we use nanoscale biology to build robots with huge potential.”

Dr. Luu said, “We’ve created a new class of nanomaterials with adjustable properties, enabling diverse applications—from adaptive materials that change optical properties in response to the environment to autonomous nanorobots designed to seek out and destroy cancer cells.”

To assemble the voxels, the team incorporate additional DNA strands on to the exterior of the nanostructures, with the new strands acting as programmable binding sites.

Dr. Luu said, “These sites act like Velcro with different colors—designed so that only strands with matching ‘colors’ (in fact, complementary DNA sequences) can connect.”

He said this innovative approach allows precise control over how voxels bind to each other, enabling the creation of customizable, highly specific architectures.

One of the most exciting applications of this technology is its potential to create nanoscale robotic boxes capable of delivering drugs directly to targeted areas within the body.

By using DNA origami, researchers can design these nanobots to respond to specific biological signals, ensuring medications are released only when and where they are needed. This targeted approach could enhance the effectiveness of cancer treatments while minimizing side effects.

In addition to drug delivery, the researchers are exploring the development of new materials that can change properties in response to environmental stimuli. For instance, these materials could be engineered to be responsive to higher loads or alter their structural characteristics based on changes in temperature or acidic (pH) levels.

Such responsive materials have the potential to transform medical, computing and electronics industries.

For more information: Science Robotics

Image: Dr. Minh Luu aligning and focusing an image on the Sydney Microscopy and Microanalysis transmission electron microscope to view a DNA origami nanostructure. Credit: Stefanie Zingsheim/University of Sydney

University of Illinois Chicago students write the book on automating diamond membrane creation for quantum devices

Six undergraduates at the University of Illinois Chicago (UIC) have been developing a process to accelerate the creation of diamond membranes, which are crucial for hosting qubits, the fundamental units of quantum information. This work is part of the research at Q-NEXT, a U.S. Department of Energy National Quantum Information Science Research Center led by Argonne National Laboratory. During their 10-week internship at Argonne, the students wrote software to automate a labor-intensive part of diamond-membrane production, finding the experience both challenging and rewarding.

Their work is enabled through Break Through Tech Chicago, an initiative that provides women and nonbinary people with internship opportunities in science and technology. Argonne staff scientist Nazar Delegan, a Q-NEXT collaborator, and UIC professor Dale Reed led the student team.

Quantum information technologies are expected to revolutionize areas such as logistics, drug development and navigation in the coming decades. Diamond membranes are a new material for hosting qubits, the core of quantum devices. The membranes have desirable properties for quantum information processing, and they open paths for integrating quantum materials with current information technologies.

Scientists are investigating the most effective ways to fabricate diamond membranes. One of the production steps — a specific process in the etching stage — requires up to 60 minutes of continual human effort and supervision.

The task before the UIC students: Put that etching process on the path to full automation.

“There are factors that can disrupt the etching process. Someone has to constantly be checking that it’s being done right,” said Fernanda Villalpando, an information decision sciences senior and the group’s project manager. ​“So we worked to automate it.”

By demonstrating proof of concept, the students laid the groundwork for the procedure so that future researchers can scale it up to industry production levels.

The membranes are created by embedding a layer of graphite between two layers of diamond. The thick bottom diamond layer serves as a platform. The tissue-thin top layer — 100 to 1,000 nanometers thin, a hundred to a thousand times thinner than a sheet of paper — is the diamond membrane. Scientists use electrical probes to chemically etch away the graphite beneath the membrane, which can then be peeled off and integrated into a quantum device.

Currently, a human must watch over the roughly hour-long etching process to ensure its successful execution. But following the UIC group’s work, researchers will one day be able to say goodbye to human-supervised etching.

Building on image detection software called Open CV as part of the Python programming language, the students created a program to teach the computer to visually assess and respond to the etching process. Is there a bubble trapped between layers? An unexpected obstruction? With the UIC group’s program, the computer knows whether to stop the etch, continue or work around it.

“That way, the scientists don’t have to be there to push the ​‘off’ button, for example,” Villalpando said. ​“Our program stops it for them.”

As the ones spearheading the procedure, the team had no blueprint for how to proceed. They quickly realized they’d have to draw heavily on their computer science knowledge, hunt for relevant documentation and even pick up the phone to call the device’s manufacturer for minutiae not captured in the literature.

“We had to reach out to the company. It was a little frustrating, because how were we going to do the rest of the work if we’re having trouble communicating with the devices?” said Claudia Jimenez, a computer science junior. ​“But once we got that part, we had the persistence and resilience to keep going, and made a lot of progress in two or three weeks. We kept going and looked for different resources to accomplish something that none of us had ever done before.”

In fact, it was something no one had done before. Currently, only a select few groups in the world are creating diamond-membrane qubit platforms.

“I love being in a space where everyone is excited about it,” Villalpando said. ​“I’ve been in rooms where people do the work that they do all the time. It’s not new, and there’s only one way to do it. But we get to be creative and think, ​‘How can we solve this?’”

For Q-NEXT, the group’s development of a technical procedure from scratch was a crucial contribution to quantum materials fabrication. For the students, it was part of the real-world work of experimenting in a laboratory.

The Chicago Quantum Exchange honored the group’s work with the Best Undergraduate Student Poster Award at the Chicago Quantum Summit in October.

The UIC team was also excited to be part of game-changing research that could have impacts across so many areas of everyday life.

“Quantum applies to so many different applications and fields and industries. I would possibly like to be a part of that. It was nice to hear from actual professionals in the field giving an explanation about what quantum is, how it can be applied and how we’re actually going to do it,” Jimenez said.

For more information: Q-NEXT

Image: UIC students work at the Argonne Quantum Foundry through the Break Through Tech Chicago program, helping automate an important step in the production of diamond membranes for qubits. (Image by Argonne National Laboratory.)

Deep learning streamlines identification of 2D materials

Researchers have developed a deep learning-based method that enhances the speed and accuracy of identifying and classifying two-dimensional (2D) materials using Raman spectroscopy. Traditional Raman analysis is slow and requires manual interpretation, but this new approach accelerates the development and analysis of 2D materials, which are crucial for electronics and medical technologies. Lead researcher Yaping Qi from Tohoku University explains that their generative model improves limited and unevenly distributed spectral data, effectively filling in the gaps.

The learning model used spectral data from seven different 2D materials and three distinct stacked combinations. The researchers introduced an innovative data augmentation framework using Denoising Diffusion Probabilistic Models (DDPM) to generate additional synthetic data and address these challenges. For this type of model, noise is added to the original data to enhance the dataset, and then the model learns to work backward and remove this noise to generate a novel output that is consistent with the original data distribution.
By pairing this augmented dataset with a four-layer Convolutional Neural Network (CNN), the research team achieved a classification accuracy of 98.8% on the original dataset and, notably, 100% accuracy with the augmented data. This automated approach not only enhances classification performance but also reduces the need for manual intervention, improving the efficiency and scalability of Raman spectroscopy for 2D material identification.

“This method provides a robust and automated solution for high-precision analysis of 2D materials,” summarizes Qi, “The integration of deep learning techniques holds significant promise for materials science research and industrial quality control, where reliable and rapid identification is critical.”

The study presents the first application of DDPM in Raman spectral data generation, paving the way for more efficient, automated spectroscopy analysis. This approach enables precise material characterization even when experimental data is scarce or difficult to obtain. Ultimately, this can allow for research done in the lab to transform into a real product that consumers can buy in stores into a much smoother process.

For more information: Tohoku University

Invisible touch: Stevens is giving AI the ability to feel and measure surfaces

AI technologies have advanced in seeing, conversing, calculating, and creating, but they have struggled to measure or “feel” surfaces. According to Stevens physics professor Yong Meng Sua, while AI has developed a sense of sight, it hasn’t yet achieved a human-like sense of touch to distinguish textures. However, researchers at Stevens’ Center for Quantum Science and Engineering (CQSE) have now demonstrated a method to give AI the ability to feel.

Sua, working with CQSE Director Yuping Huang and doctoral candidates Daniel Tafone and Luke McEvoy ’22 M.S. ‘23, devised a quantum-lab setup that combines a photon-firing scanning laser with new algorithmic AI models trained to tell the differences among various surfaces as they are imaged with those lasers.

In their system, a specially created beam of light is pulsed in short blasts at a surface to “feel” it. Reflected, back-scattered photons return from the target object carrying speckle noise, a random type of flaw that occurs in imagery.

Speckle noise is normally considered detrimental to clear, accurate imaging. However, the Stevens group’s system takes a different approach: it detects and processes these noise artifacts using an AI that has been carefully trained to interpret their characteristics as valuable data. This allows the system to accurately discern the topography of the object.

“We use the variation in photon counts over different illumination points across the surface,” says Tafone.

The team used 31 industrial sandpapers with surfaces of varying roughness, ranging from 1 to 100 microns thick, as experimental targets. (For comparison, an average human hair is about 100 microns thick.) Mode-locked lasers generated light pulses aimed at the samples.

Those pulses passed through transceivers, encountered the sandpapers, and then rebounded back through the system for analysis by the team’s learning model.

During early tests, the group’s method averaged a root-mean-square error (RMSE) of about 8 microns; after working with multiple samples and averaging results across them, its accuracy improved significantly to within 4 microns, comparable to the best industrial profilometer devices currently used.

“Interestingly, our system worked best for the finest-grained surfaces, such as diamond lapping film and aluminum oxide,” notes Tafone.

The new method could be useful for various applications, he adds.

For example, human examiners often make mistakes when attempting to detect skin cancers, confusing very similar-looking but harmless conditions with potentially fatal melanomas.

“Tiny differences in mole roughness, too small to see with the human eye but measurable with our proposed quantum system, could differentiate between those conditions,” explains Huang. “Quantum interactions provide a wealth of information, using AI to quickly understand and process it is the next logical step.”

Manufacturing quality control of components, as well, often hinges on extremely small distances that can mean the difference between a perfect part and a tiny defect that could eventually cause a dangerous mechanical failure.

“Since LiDAR technology is already implemented widely in devices such as autonomous cars, smartphones and robots,” Huang concludes, “our method enriches their capabilities with surface property measurement at very small scales.”

For more information: Applied Optics

New Machine Learning Model Predicts Dielectric Function of Materials

Cutting-edge researchers Tomohito Amano and Shinji Tsuneyuki at the University of Tokyo, alongside Tamio Yamazaki from CURIE, have unveiled a machine learning model that revolutionizes our approach to predicting the dielectric function of materials – moving beyond traditional first-principles calculations.

This critical function gauges the polarization of negative and positive charges within materials, forming the core concept behind dielectric materials.

With this innovative model, scientists can achieve rapid and precise predictions of dielectric functions, paving the way for the creation of next-generation dielectric materials essential for transformative technologies, including the anticipated advancements in 6G network.

While not as widely recognized as semiconductors, dielectric materials hold immense promise for enhancing modern electronic systems. These materials offer a unique characteristic: they do not conduct electricity well but are not insulators either. When subjected to an electric field, positive charges within the material are drawn toward the field, while negative charges shift in the opposite direction, leading to dielectric polarization.

This polarization is quantified by the dielectric function, a critical measure of its strength. Despite its significance, calculating the dielectric function generally requires complex first-principles approaches rooted in quantum mechanics, making it a time-consuming and resource-intensive endeavor.

Researchers have successfully developed an innovative machine-learning model designed to tackle significant challenges in materials science. They’ve taken a groundbreaking approach by generating training data through first-principle calculations of the electronic states of various materials.

Instead of relying on conventional calculations focused on individual molecules, this model emphasizes the chemical bonds between atoms. Its accuracy was rigorously tested against empirical data from simple molecules like methanol and ethanol.

This model not only describes the electronic states of a variety of materials with nearly the same accuracy as traditional methods but also does so while dramatically reducing computational demands. Its capability for large-scale and long-term simulations opens new doors, enabling researchers to investigate the macroscopic origins of dielectric properties in complex molecular systems—a feat previously hindered by computational costs.

Despite these impressive achievements, scientists are already envisioning future possibilities.

For more information: Physical Review B