Computers Math

An international team of researchers has designed and built a chip that runs computations directly in memory and can run a wide variety of AI applications-all at a fraction of the energy consumed by computing platforms for general-purpose AI computing.
The NeuRRAM neuromorphic chip brings AI a step closer to running on a broad range of edge devices, disconnected from the cloud, where they can perform sophisticated cognitive tasks anywhere and anytime without relying on a network connection to a centralized server. Applications abound in every corner of the world and every facet of our lives, and range from smart watches, to VR headsets, smart earbuds, smart sensors in factories and rovers for space exploration.
The NeuRRAM chip is not only twice as energy efficient as the state-of-the-art “compute-in-memory” chips, an innovative class of hybrid chips that runs computations in memory, it also delivers results that are just as accurate as conventional digital chips. Conventional AI platforms are a lot bulkier and typically are constrained to using large data servers operating in the cloud.
In addition, the NeuRRAM chip is highly versatile and supports many different neural network models and architectures. As a result, the chip can be used for many different applications, including image recognition and reconstruction as well as voice recognition.
“The conventional wisdom is that the higher efficiency of compute-in-memory is at the cost of versatility, but our NeuRRAM chip obtains efficiency while not sacrificing versatility,” said Weier Wan, the paper’s first corresponding author and a recent Ph.D. graduate of Stanford University who worked on the chip while at UC San Diego, where he was co-advised by Gert Cauwenberghs in the Department of Bioengineering.
The research team, co-led by bioengineers at the University of California San Diego, presents their results in the Aug. 17 issue of Nature. More

Researchers from Göttingen (Germany) and Lausanne (Switzerland) have successfully created electron-photon pairs in a controlled way in an electron microscope for the first time. Using a new method, they could precisely detect the involved particles. The findings of the study expand the toolbox of quantum technology.
Faster computers, tap-proof communication, better car sensors — quantum technologies have the potential to revolutionize our lives just as once the invention of computers or the internet. Experts worldwide are trying to implement findings from basic research into quantum technologies. To this end, they often require individual particles, such as photons — the elementary particles of light — with tailored properties. However, obtaining individual particles is complicated and requires intricate methods. In a study recently published in the journal Science, researchers now present a new method that simultaneously generates two individual particles in form of a pair.
The international team from the Göttingen Max Planck Institute (MPI) for Multidisciplinary Sciences, the University of Göttingen, and the Swiss Federal Institute of Technology in Lausanne (EPFL) succeeded in coupling single free electrons and photons in an electron microscope. In the Göttingen experiment, the beam from an electron microscope passes through an integrated optical chip, fabricated by the Swiss team. The chip consists of a fiber-optic coupling and a ring-shaped resonator that stores light by keeping moving photons on a circular path. “When an electron scatters at the initially empty resonator, a photon is generated,” explains Armin Feist, scientist at the MPI and one of the study’s first authors. “In the process, the electron loses exactly the amount of energy that the photon requires to be created virtually from nothing in the resonator. As a result, the two particles are coupled through their interaction and form a pair.” With an improved measurement method, the physicists could precisely detect the individual particles involved and their simultaneous manifestation.
Future quantum technology with free electrons
“With the electron-photon pair, we only need to measure one particle to obtain information about the energy content and temporal appearance of the second one,” emphasizes Germaine Arend, a PhD candidate at the MPI and also first author of the study. This allows researchers to use one quantum particle in an experiment while, at the same time, confirming its presence by detecting the other particle, in a so-called heralding scheme. Such a feature is necessary for many applications in quantum technology.
Max Planck Director Claus Ropers sees electron-photon pairs as a new opportunity for quantum research: “The method opens up fascinating new possibilities in electron microscopy. In the field of quantum optics, entangled photon pairs already improve imaging. With our work, such concepts can now be explored with electrons.” Tobias Kippenberg, professor at EPFL adds, “For the first time, we bring free electrons into the toolbox of quantum information science. More broadly, coupling free electrons and light using integrated photonics could open the way to a new class of hybrid quantum technologies.”
Story Source:
Materials provided by Max-Planck-Gesellschaft. Note: Content may be edited for style and length. More

Superconductors are the key to lossless current flow. However, the realization of superconducting diodes has only recently become an important topic of fundamental research. An international research team involving the theoretical physicist Mathias Scheurer from the University of Innsbruck have now succeeded in reaching a milestone: the realization of a superconducting diode effect without an external magnetic field, thus proving the assumption that superconductivity and magnetism coexist. They report on this in Nature Physics.
One speaks of a superconducting diode effect when a material behaves like a superconductor in one direction of current flow and like a resistor in the other. In contrast to a conventional diode, such a superconducting diode exhibits a completely vanishing resistance and thus no losses in the forward direction. This could form the basis for future lossless quantum electronics. Physicists first succeeded in creating the diode effect about two years ago, but with some fundamental limitations. “At that time, the effect was very weak and it was generated by an external magnetic field, which is very disadvantageous in potential technological applications,” explains Mathias Scheurer from the Institute of Theoretical Physics at the University of Innsbruck.
The new experiments carried out by experimental physicists at Brown University, described in the current issue of Nature Physics, do not require an external magnetic field. In addition to the aforementioned application-relevant advantages, the experiments confirm a thesis previously theorized by Mathias Scheurer: Namely, that superconductivity and magnetism coexist in a system consisting of three graphene layers twisted against each other. The system thus virtually generates its own internal magnetic field, creating a diode effect. “The diode effect observed by colleagues at Brown University was additionally very strong. Moreover, the diode direction can be reversed by a simple electric field. Together, this makes trilayer graphene such a promising platform for the superconducting diode effect,” clarifies Mathias Scheurer, who received an ERC Starting Grant this year for his research on two-dimensional materials, especially graphene.
Promising material graphene
The diode effect described in Nature Physics was also produced with graphene, a material consisting of a single layer of carbon atoms arranged in a honeycomb pattern. Stacking several layers of graphene leads to completely new properties, including the ability of three graphene layers twisted against each other to conduct electric current without loss. The fact that a superconducting diode effect exists without an external magnetic field in this system has great implications for the study of the complex physical behavior of twisted trilayer graphene, as it demonstrates the coexistence of superconductivity and magnetism. This shows that the diode effect not only has technological relevance, but also has the potential to improve our understanding of fundamental processes in many-body physics. The theoretical basis for this has already been published in another high-ranking publication.
Story Source:
Materials provided by University of Innsbruck. Note: Content may be edited for style and length. More

Engineers have created intelligent 3D printers that can quickly detect and correct errors, even in previously unseen designs, or unfamiliar materials like ketchup and mayonnaise, by learning from the experiences of other machines.
The engineers, from the University of Cambridge, developed a machine learning algorithm that can detect and correct a wide variety of different errors in real time, and can be easily added to new or existing machines to enhance their capabilities. 3D printers using the algorithm could also learn how to print new materials by themselves. Details of their low-cost approach are reported in the journal Nature Communications.
3D printing has the potential to revolutionise the production of complex and customised parts, such as aircraft components, personalised medical implants, or even intricate sweets, and could also transform manufacturing supply chains. However, it is also vulnerable to production errors, from small-scale inaccuracies and mechanical weaknesses through to total build failures.
Currently, the way to prevent or correct these errors is for a skilled worker to observe the process. The worker must recognise an error (a challenge even for the trained eye), stop the print, remove the part, and adjust settings for a new part. If a new material or printer is used, the process takes more time as the worker learns the new setup. Even then, errors may be missed as workers cannot continuously observe multiple printers at the same time, especially for long prints.
“3D printing is challenging because there’s a lot that can go wrong, and so quite often 3D prints will fail,” said Dr Sebastian Pattinson from Cambridge’s Department of Engineering, the paper’s senior author. “When that happens, all of the material and time and energy that you used is lost.”
Engineers have been developing automated 3D printing monitoring, but existing systems can only detect a limited range of errors in one part, one material and one printing system. More

There’s still plenty of room at the bottom to generate piezoelectricity. Engineers at Rice University and their colleagues are showing the way.
A new study describes the discovery of piezoelectricity — the phenomenon by which mechanical energy turns into electrical energy — across phase boundaries of two-dimensional materials.
The work led by Rice materials scientists Pulickel Ajayan and Hanyu Zhu and their colleagues at Rice’s George R. Brown School of Engineering, the University of Southern California, the University of Houston, Wright-Patterson Air Force Base Research Laboratory and Pennsylvania State University appears in Advanced Materials.
The discovery could aid in the development of ever-smaller nanoelectromechanical systems, devices that could be used, for example, to power tiny actuators and implantable biosensors, and ultrasensitive temperature or pressure sensors.
The researchers show the atomically thin system of a metallic domain surrounding semiconducting islands creates a mechanical response in the material’s crystal lattice when subjected to an applied voltage.
The presence of piezoelectricity in 2D materials often depends on the number of layers, but synthesizing the materials with a precise number of layers has been a formidable challenge, said Rice research scientist Anand Puthirath, co-lead author of the paper. More

When early humans discovered how to harness fire, they were able to push back against the nightly darkness that enveloped them. With the invention and widespread adoption of electricity, it became easier to separate heat from light, work through the night, and illuminate train cars to highways. In recent years, old forms of electric light generation such as halogen lightbulbs have given way to more energy efficient alternatives, further cheapening the costs to brighten our homes, workplaces, and lives generally.
Unfortunately, however, white light generation by newer technologies such as light-emitting diodes (LEDs) is not straightforward and often relies on a category of materials called “rare-earth metals,” which are increasingly scarce. This has recently led scientists to look for ways to produce white light more sustainably. Researchers at Giessen University, the University of Marburg, and Karlsruhe Institute of Technology have recently uncovered a new class of material called a “cluster glass” that shows great potential for replacing LEDs in many applications.
“We are witnessing the birth of white-light generation technology that can replace current light sources. It brings all the requirements that our society asks for: availability of resources, sustainability, biocompatibility,” said Prof. Dr. Simone Sanna, Giessen University Professor and lead computational researcher on the project. “My colleagues from the experimental sciences, who observed this unexpected white light generation, asked for theoretical support. Cluster glass has an incredible optical response, but we don’t understand why. Computational methods can help to understand those mechanisms. This is exactly the challenge that theoreticians want to face.”
Sanna and his collaborators have turned to the power of high-performance computing (HPC), using the Hawk supercomputer at the High-Performance Computing Center Stuttgart (HLRS) to better understand cluster glass and how it might serve as a next-generation light source. They published their findings in Advanced Materials.
Clear-eyed view on cluster glass formation
If you are not a materials scientist or chemist, the word glass might just mean the clear, solid material in your windows or on your dinner table. Glass is actually a class of materials that are considered “amorphous solids;” that is, they lack an ordered crystalline lattice, often due to a rapid cooling process. At the atomic level, their constituent particles are in a suspended, disordered state. Unlike crystal materials, where particles are orderly and symmetrical across a long molecular distance, glasses’ disorder at the molecular level make them great for bending, fragmenting, or reflecting light. More

By using photons and electron spin qubits to control nuclear spins in a two-dimensional material, researchers at Purdue University have opened a new frontier in quantum science and technology, enabling applications like atomic-scale nuclear magnetic resonance spectroscopy, and to read and write quantum information with nuclear spins in 2D materials.
As published Monday (Aug. 15) in Nature Materials, the research team used electron spin qubits as atomic-scale sensors, and also to effect the first experimental control of nuclear spin qubits in ultrathin hexagonal boron nitride.
“This is the first work showing optical initialization and coherent control of nuclear spins in 2D materials,” said corresponding author Tongcang Li, a Purdue associate professor of physics and astronomy and electrical and computer engineering, and member of the Purdue Quantum Science and Engineering Institute.
“Now we can use light to initialize nuclear spins and with that control, we can write and read quantum information with nuclear spins in 2D materials. This method can have many different applications in quantum memory, quantum sensing, and quantum simulation.”
Quantum technology depends on the qubit, which is the quantum version of a classical computer bit. It is often built with an atom, subatomic particle, or photon instead of a silicon transistor. In an electron or nuclear spin qubit, the familiar binary “0” or “1” state of a classical computer bit is represented by spin, a property that is loosely analogous to magnetic polarity — meaning the spin is sensitive to an electromagnetic field. To perform any task, the spin must first be controlled and coherent, or durable.
The spin qubit can then be used as a sensor, probing, for example, the structure of a protein, or the temperature of a target with nanoscale resolution. Electrons trapped in the defects of 3D diamond crystals have produced imaging and sensing resolution in the 10-100 nanometer range. More

An international research team led by the University of Göttingen has detected novel quantum effects in high-precision studies of natural double-layer graphene and has interpreted them together with the University of Texas at Dallas using their theoretical work. This research provides new insights into the interaction of the charge carriers and the different phases, and contributes to the understanding of the processes involved. The LMU in Munich and the National Institute for Materials Science in Tsukuba, Japan, were also involved in the research. The results were published in Nature.
The novel material graphene, a wafer-thin layer of carbon atoms, was first discovered by a British research team in 2004. Among other unusual properties, graphene is known for its extraordinarily high electrical conductivity. If two individual graphene layers are twisted at a very specific angle to each other, the system even becomes superconducting, i.e. conducts electricity without any resistance, and exhibits other exciting quantum effects such as magnetism. However, the production of such twisted graphene double-layers has so far required increased technical effort.
This novel study used the naturally occurring form of double-layer graphene, where no complex fabrication is required. In a first step, the sample is isolated from a piece of graphite in the laboratory using a simple adhesive tape. To observe quantum mechanical effects, the Göttingen team then applied a high electric field perpendicular to the sample: the electronic structure of the system changes and a strong accumulation of charge carriers with similar energy occurs.
At temperatures just above absolute zero of minus 273.15 degrees Celsius, the electrons in the graphene can interact with each other — and a variety of complex quantum phases emerge completely unexpectedly. For example, the interactions cause the spins of the electrons to align, making the material magnetic without any further external influence. By changing the electric field, researchers can continuously change the strength of the interactions of the charge carriers in the double-layer graphene. Under specific conditions, the electrons can be so restricted in their freedom of movement that they form their own electron lattice and can no longer contribute to transporting charge due to their mutual repulsive interaction. The system is then electrically insulating.
“Future research can now focus on investigating further quantum states,” say Professor Thomas Weitz and PhD student Anna Seiler, Faculty of Physics at Göttingen University. “In order to access other applications, for example novel computer systems such as quantum computers, researchers would need to find how these results could be achieved at higher temperatures. However, a major advantage of the current system developed in our new research lies in the simplicity of the fabrication of the materials.”
Story Source:
Materials provided by University of Göttingen. Note: Content may be edited for style and length. More