Deep learning-enabled prediction of 2D material breakdown.

Characterizing electrical breakdown limits of materials is a crucial step in device development. However, methods for repeatable measurements are scarce in two-dimensional materials, where breakdown studies have been limited to destructive methods. This restricts our ability to fully account for variability in local electronic properties induced by surface contaminants and the fabrication process. To tackle this, we implement a two-step deep-learning model to predict the breakdown mechanism and breakdown voltage of monolayer MoS2devices with varying channel lengths and resistances using current measured in the low-voltage regime as inputs. A deep neural network (DNN) first classifies between Joule and avalanche breakdown mechanisms using partial current traces from 0 to 20 V. Following this, a convolutional long short-term memory network (CLSTM) predicts breakdown voltages of these classified devices based on partial current traces. We test our model with electrical measurements collected using feedback-control of the applied voltage to prevent device destruction, and show that the DNN classifier achieves an accuracy of 79% while the CLSTM model has a 12% error when requiring only 80% of the current trace as inputs. Our results indicate that information encoded in the current behavior far from the breakdown point can be used for breakdown predictions, which will enable non-destructive and rapid material characterization for 2D material device development.

Control and single-shot readout of an ion embedded in a nanophotonic cavity.

Distributing entanglement over long distances using optical networks is an intriguing macroscopic quantum phenomenon with applications in quantum systems for advanced computing and secure communication1,2. Building quantum networks requires scalable quantum light-matter interfaces1 based on atoms3, ions4 or other optically addressable qubits. Solid-state emitters5, such as quantum dots and defects in diamond or silicon carbide6-10, have emerged as promising candidates for such interfaces. So far, it has not been possible to scale up these systems, motivating the development of alternative platforms. A central challenge is identifying emitters that exhibit coherent optical and spin transitions while coupled to photonic cavities that enhance the light-matter interaction and channel emission into optical fibres. Rare-earth ions in crystals are known to have highly coherent 4f-4f optical and spin transitions suited to quantum storage and transduction11-15, but only recently have single rare-earth ions been isolated16,17 and coupled to nanocavities18,19. The crucial next steps towards using single rare-earth ions for quantum networks are realizing long spin coherence and single-shot readout in photonic resonators. Here we demonstrate spin initialization, coherent optical and spin manipulation, and high-fidelity single-shot optical readout of the hyperfine spin state of single 171Yb3+ ions coupled to a nanophotonic cavity fabricated in an yttrium orthovanadate host crystal. These ions have optical and spin transitions that are first-order insensitive to magnetic field fluctuations, enabling optical linewidths of less than one megahertz and spin coherence times exceeding thirty milliseconds for cavity-coupled ions, even at temperatures greater than one kelvin. The cavity-enhanced optical emission rate facilitates efficient spin initialization and single-shot readout with conditional fidelity greater than 95 per cent. These results showcase a solid-state platform based on single coherent rare-earth ions for the future quantum internet.