EDITORS' SUGGESTION
Defective devices can severely impact the performance of hardware-based neural networks, in particular resistive crossbar arrays. This study introduces a network training approach that reduces the influence of defective devices, maintaining inference accuracy. The authors demonstrate this approach on a set of dies each containing a crossbar array consisting of 20,000 magnetic tunnel junction devices. They also develop a generalized approach using the statistics of defects and demonstrate similar performance on all dies. These results translate to a manufacturing setting where millions of dies with possible defects are produced, but the performance of even subpar chips can be guaranteed.
William A. Borders et al.
Phys. Rev. Applied 21, 054028 (2024)
LETTER
Ramsey interferometry is an important technique in precision spectroscopy and quantum coherence measurement. The authors explore an innovative scheme in which splitter pulses are implemented by geometrical means, eliminating the temporal dependence of the atom-light interaction. This translates to an interferometer that is insensitive to the mean velocity of the atomic ensemble, making it suitable for applications in quantum computing and simulation, as well as atomtronic circuits. Using this geometric Ramsey interferometer, the team measures the phase accumulation during the free-evolution time due to a geometric scalar term.
Chetan Sriram Madasu, Ketan Damji Rathod, Chang Chi Kwong, and David Wilkowski
Phys. Rev. Applied 21, L051001 (2024)
EDITORS' SUGGESTION
Transition-metal dichalcogenides (TMDCs) are promising building blocks for future electronic circuits, but their performance is often hindered by poorly understood electron-phonon interactions. This study leverages a fresh ab initio approach, combining density-functional theory with the linearized Boltzmann transport equation (LBTE) and nonequilibrium Green’s functions (NEGF), to explore phonon-limited transport in TMDCs. The authors find that LBTE and NEGF return very similar mobility values despite the different approximations upon which they rely, thus paving the way for comprehensive device simulations that include electron-phonon scattering.
Jonathan Backman, Youseung Lee, and Mathieu Luisier
Phys. Rev. Applied 21, 054017 (2024)
LETTER
Accurate measurement of a material’s coefficient of thermal expansion is a potentially powerful probe for investigating phase transitions. Unfortunately, familiar techniques are either low-resolution or put stringent requirements on sample preparation. This study implements ultrastable optical-fiber interferometry for contactless optical measurement with picometer spatial resolution, and can measure fragile samples less than 100 µm thick with millikelvin-level temperature resolution. Using their method on BaFeAs, the authors discover hysteresis at the first-order phase transition between its antiferromagnetic and paramagnetic phases, with a boundary moving at about 188 µm/s.
Xin Qin et al.
Phys. Rev. Applied 21, L041003 (2024)
PERSPECTIVE
Spin waves and their quanta magnons are the collective excitations of a spin systems of a magnetic material, which offer the potential for higher efficiency and lower energy consumption in solving specific issues in data processing. This Perspective discusses the current challenges in realizing magnonic circuits based on the building blocks developed to date, and further looks at the application of magnons in neuromorphic networks and stochastic, reservoir, and quantum computing, and discusses their advantages over conventional electronics in these areas.
Qi Wang et al.
Phys. Rev. Applied 21, 040503 (2024)
EDITORS' SUGGESTION
Machines based on coupled bistable oscillators can rapidly produce high-quality solutions to difficult problems in combinatorial optimization. While the dynamics of such systems can be derived, exactly why these dynamics are so good for optimization is unclear. This study presents a complete mathematical equivalence between coupled-oscillator machines and the primal-dual method of Lagrange multipliers, elucidating the precise mathematical role of each hardware component and enabling the principled design of more sophisticated optimization machines. Simulations show that such a circuit consumes extremely low amounts of power and energy per optimization, even for many variables.
Sri Krishna Vadlamani, Tianyao Patrick Xiao, and Eli Yablonovitch
Phys. Rev. Applied 21, 044042 (2024)
LETTER
While haloscope experiments searching for axion dark matter with cylindrical microwave cavity resonators are the most sensitive to date, that sensitivity is degraded at high frequencies, due to geometric scaling. The authors demonstrate a prototype thin-shell cavity resonator that decouples volume from resonant frequency, and thus avoids such degradation. As the resonator comprises two mechanically isolated pieces, a protocol is developed for robust, automated precision alignment, which enables a wide tuning range for the resonator’s axion-sensitive TM mode. A discussion of the instrument’s feasibility for high-frequency probes of the post-inflationary scenario is also offered.
Taj A. Dyson et al.
Phys. Rev. Applied 21, L041002 (2024)
EDITORS' SUGGESTION
Microwave-free magnetometry with N- centers has emerged as a complementary method to traditional techniques when the use of microwaves is impractical, particularly in applications involving metals and biological samples. Integration of this method with imaging capabilities offers the potential for nondestructive probing in a 2D spatial plane, while also capturing temporal dynamics within existing technological constraints. It is evident that the limits of sensitivity have not been fully realized, and improvements may be achieved via faster specialized cameras and advanced color-center fabrication.
Joseph Shaji Rebeirro et al.
Phys. Rev. Applied 21, 044039 (2024)
PERSPECTIVE
Dynamic beamforming is critical in applications such as radar detection, holographic imaging, and reconfigurable intelligent surfaces (RIS). This Perspective reviews a revolutionary and economical technique to achieve dynamic beamforming, utilizing the moiré pattern formed by twisted stacked metasurfaces. Research here faces challenges such as far-field calculations and the inverse design of specific radiation patterns, due to our limited understanding of the complex mode coupling between the moiré pattern and the metallic back plate. The authors outline potential solutions and project the future applications and research directions for the reflective moiré metasurface.
Shuo Liu and Tie Jun Cui
Phys. Rev. Applied 21, 040502 (2024)