Entangled states of matter are a crucial resource for many quantum tasks, but producing such states deterministically between noninteracting parties is quite challenging, especially in solid-state systems. Here researchers offer a protocol to entangle two ensembles of nuclear spins that surround two quantum dots connected by an electron waveguide, employing sequential interaction of the nuclei with spin-polarized electrons. The authors numerically demonstrate the efficacy of the protocol, even in disordered and noisy setups, which suggests its feasibility with state-of-the-art techniques in quantum information processing.

A lively discussion has emerged in both the valleytronics and spintronics communities, concerning the correct interpretation of nonlocal transport experiments. The authors contribute to this ongoing debate their overview of all mechanisms that can create charge-induced nonlocal voltages, which could be misattributed to spin- or valley-related effects. A detailed description of a further measurement artifact, apparently not discussed in the literature, is also included. Fortunately, special instrumentation can significantly reduce most of these measurement artifacts, and thus can strongly diminish the risk of misinterpretation.

Hardware neural networks based on synaptic devices can provide the speed, parallelism, and low energy consumption needed for applications in distributed computing and the Internet of Things. However, the accuracy of inference in a given system can be severely impacted by device variations and other imperfections. In this work, the authors demonstrate that a binary neural network using a passive crossbar array of magnetic tunnel junctions can perform well on a simple wine identification dataset. This work demonstrates the promise of these devices but cautions that careful consideration of hardware realities in scaling up this technology is necessary for real-world applications.

Leveraging physical processes rather than a central processor is key to building machine learning systems that are massively scalable, robust to damage, and energy-efficient, like the brain. To achieve these features, the authors build an electrical network made of identical resistive edges that self-adjust based on local conditions in order to minimize an energy-based global cost function when shown training examples. Problems like regression and data classification are successfully solved by this network. Due to their energy efficiency and scaling advantages, future versions may one day compete with computational neural networks.

Quantum diamond microscopy is emerging as a tool for magnetic current imaging (MCI) of electrical devices. Due to the optical invasiveness of the technique, it has not been extended to photovoltaic (PV) devices—until now. This study employs a stroboscopic measurement protocol that allows temporally resolved measurements of silicon PV cells, independent of stimulus. The authors capitalize on the contactless aspect of MCI to image photocurrent loops that are entirely internal to the device (e.g. under open-circuit conditions), equivalent to shunt-resistance paths. The versatile capabilities of quantum diamond microscopy may find wide application in this further context.

Although there has been extensive interest in combining organic and two-dimensional (2D) materials to produce hybrid heterostructures for optoelectronic applications (including photovoltaics), very few studies have demonstrated how such hybrid structures might outperform structures formed by either component alone. This work presents a scalable method to produce centimeter-sized organic-2D multilayer structures, and shows that the photon-to-free-carrier conversion yield is significantly enhanced when monolayer h-BN is inserted at an organic donor-acceptor interface. This result should inspire future studies on incorporating 2D materials into organic devices to improve efficiency.

The analytical modeling of electrically driven microbeams is essential to designing the silicon sensors and actuators behind much of the functionality in smartphones, for example. Such modeling is a challenge, because Coulomb forces acting on elastic structures typically yield complex dynamical behavior. This study applies ideas regarding the Central Limit Theorem of analytical probability theory to evaluate relevant Coulomb integrals, leading to a surprisingly accurate model with a single degree of freedom. The approach can presumably be generalized beyond the context of micromachines, to the calculation of a broad class of integrals with kernels close to a singularity.

In one branch of quantum information processing, high cooperativity between spins and a resonator is essential. Planar superconducting resonators present an opportunity for miniaturization and fully integrated circuitry, but unfortunately their cooperativity is a fraction of that of their bulk counterparts. This study designs a planar microwave resonator to achieve homogeneous magnetic field, and utilizes the anisotropic g-factor tensor of erbium spins in yttrium orthosilicate for maximum spin-photon coupling. Cooperativity on par with that of a three-dimensional cavity is achieved, marking an exciting step toward integrated circuits for spin quantum information processing.

The application of spintronics to physical reservoir computing has great potential, but development still suffers from inevitable technical complications in nanofabrication. This numerical study considers spin waves excited in a self-organized skyrmion lattice under a magnetic field in a chiral magnet—which would not require advanced manufacturing in practice—to make progress on the problem. Such a skyrmion lattice offers great levels of generalizability, memory capacity, and nonlinearity, fulfilling the fundamental requirements of reservoir computing. The results will promote engineering solutions to pave the way toward reliable, energy-conserving reservoir computing.

More than a century ago, researchers observed that colloidal particles attached to bubbles and drops stabilize foams and emulsions, suggesting that particle hydrophobicity is in control. This trick is difficult to implement at industrial scale, though, and its physical processes are still debated. This Letter presents an encapsulation technique that is both practical and quantitatively described by heterogeneous electrostatic double-layer interactions, not hydrophobicity. The authors create—at low cost and industrial scale—a wide variety of “bubbloons” and “droploids”, thanks to diverse protective shells that remain stable for years. This could impact a host of industrial applications.

Entangled states of matter are a crucial resource for many quantum tasks, but producing such states deterministically between noninteracting parties is quite challenging, especially in solid-state systems. Here researchers offer a protocol to entangle two ensembles of nuclear spins that surround two quantum dots connected by an electron waveguide, employing sequential interaction of the nuclei with spin-polarized electrons. The authors numerically demonstrate the efficacy of the protocol, even in disordered and noisy setups, which suggests its feasibility with state-of-the-art techniques in quantum information processing.

A lively discussion has emerged in both the valleytronics and spintronics communities, concerning the correct interpretation of nonlocal transport experiments. The authors contribute to this ongoing debate their overview of all mechanisms that can create charge-induced nonlocal voltages, which could be misattributed to spin- or valley-related effects. A detailed description of a further measurement artifact, apparently not discussed in the literature, is also included. Fortunately, special instrumentation can significantly reduce most of these measurement artifacts, and thus can strongly diminish the risk of misinterpretation.

Hardware neural networks based on synaptic devices can provide the speed, parallelism, and low energy consumption needed for applications in distributed computing and the Internet of Things. However, the accuracy of inference in a given system can be severely impacted by device variations and other imperfections. In this work, the authors demonstrate that a binary neural network using a passive crossbar array of magnetic tunnel junctions can perform well on a simple wine identification dataset. This work demonstrates the promise of these devices but cautions that careful consideration of hardware realities in scaling up this technology is necessary for real-world applications.

Leveraging physical processes rather than a central processor is key to building machine learning systems that are massively scalable, robust to damage, and energy-efficient, like the brain. To achieve these features, the authors build an electrical network made of identical resistive edges that self-adjust based on local conditions in order to minimize an energy-based global cost function when shown training examples. Problems like regression and data classification are successfully solved by this network. Due to their energy efficiency and scaling advantages, future versions may one day compete with computational neural networks.

Quantum diamond microscopy is emerging as a tool for magnetic current imaging (MCI) of electrical devices. Due to the optical invasiveness of the technique, it has not been extended to photovoltaic (PV) devices—until now. This study employs a stroboscopic measurement protocol that allows temporally resolved measurements of silicon PV cells, independent of stimulus. The authors capitalize on the contactless aspect of MCI to image photocurrent loops that are entirely internal to the device (e.g. under open-circuit conditions), equivalent to shunt-resistance paths. The versatile capabilities of quantum diamond microscopy may find wide application in this further context.

Although there has been extensive interest in combining organic and two-dimensional (2D) materials to produce hybrid heterostructures for optoelectronic applications (including photovoltaics), very few studies have demonstrated how such hybrid structures might outperform structures formed by either component alone. This work presents a scalable method to produce centimeter-sized organic-2D multilayer structures, and shows that the photon-to-free-carrier conversion yield is significantly enhanced when monolayer h-BN is inserted at an organic donor-acceptor interface. This result should inspire future studies on incorporating 2D materials into organic devices to improve efficiency.

The analytical modeling of electrically driven microbeams is essential to designing the silicon sensors and actuators behind much of the functionality in smartphones, for example. Such modeling is a challenge, because Coulomb forces acting on elastic structures typically yield complex dynamical behavior. This study applies ideas regarding the Central Limit Theorem of analytical probability theory to evaluate relevant Coulomb integrals, leading to a surprisingly accurate model with a single degree of freedom. The approach can presumably be generalized beyond the context of micromachines, to the calculation of a broad class of integrals with kernels close to a singularity.

Developing sustainable and stable visible-light-driven photocatalysts to convert solar energy by splitting water into H${}_2$ and O${}_2$ is a great challenge in applied research. The authors’ calculations show that two-dimensional BCN is a very promising semiconductor for optoelectronic applications. Its excellent stability, sizable band gap, high carrier mobility, and distinct excitonic peaks are promising for photovoltaics, with its valence and conduction bands ideally straddling the oxidation and reduction potentials of water. Its optical absorption coefficient in the visible and near-ultraviolet regions is comparable to that of the perovskites currently employed in solar cells.

In one branch of quantum information processing, high cooperativity between spins and a resonator is essential. Planar superconducting resonators present an opportunity for miniaturization and fully integrated circuitry, but unfortunately their cooperativity is a fraction of that of their bulk counterparts. This study designs a planar microwave resonator to achieve homogeneous magnetic field, and utilizes the anisotropic g-factor tensor of erbium spins in yttrium orthosilicate for maximum spin-photon coupling. Cooperativity on par with that of a three-dimensional cavity is achieved, marking an exciting step toward integrated circuits for spin quantum information processing.

The application of spintronics to physical reservoir computing has great potential, but development still suffers from inevitable technical complications in nanofabrication. This numerical study considers spin waves excited in a self-organized skyrmion lattice under a magnetic field in a chiral magnet—which would not require advanced manufacturing in practice—to make progress on the problem. Such a skyrmion lattice offers great levels of generalizability, memory capacity, and nonlinearity, fulfilling the fundamental requirements of reservoir computing. The results will promote engineering solutions to pave the way toward reliable, energy-conserving reservoir computing.

Memristors have great potential in next-generation smart electronics and neuromorphic computing. The halide perovskites (HP) receive increasing attention in this context, due to their mixed ionic‐electronic conduction behavior, adjustable band gap, facile fabrication, low operating current, and ultralow leakage current. Advanced synaptic functions from HP memristors could be realized via more complex, three-terminal device architectures with low energy consumption. This review provides detailed insight into the operating mechanism, recent advancements, and remaining challenges of HP memristors, and suggests future prospects for the development of next-generation neuromorphic devices.