Two-dimensional (2D) materials and their heterostructures, with wafer-scale synthesis methods and fascinating properties, have attracted significant interest and triggered revolutions in corresponding device applications. However, facile methods to realize accurate, intelligent, and large-area characterizations of these 2D nanostructures are still highly desired. Herein, we report the successful application of machine-learning strategy in the optical identification of 2D nanostructures. The machine-learning optical identification (MOI) method endows optical microscopy with intelligent insight into the characteristic color information of 2D nanostructures in the optical photograph. The experimental results indicate that the MOI method enables accurate, intelligent, and large-area characterizations of graphene, molybdenum disulfide, and their heterostructures, including identifications of the thickness, existence of impurities, and even stacking order. With the convergence of artificial intelligence and nanoscience, this intelligent identification method can certainly promote fundamental research and wafer-scale device applications of 2D nanostructures. Open image in new window

Significance: Spintronic devices have become promising candidates for next-generation memory architecture. However, state-of-the-art devices, such as perpendicular magnetic tunnel junctions (MTJs), are still fundamentally constrained by a subnanosecond speed limitation, which has remained a long-lasting scientific obstacle in the ultrafast spintronics field. The highlight of our work is the demonstration of an optospintronic tunnel junction, an all-optical MTJ device which emerges as a new category of integrated photonic–spintronic memory. We demonstrate 1) laser-induced deterministic and efficient writing by an all-optical approach and electrical readout by tunnel magnetoresistance, 2) writing speed within 10 ps, demonstrated by femtosecond-resolved measurements, and 3) integration with state-of-the-art MTJ performance and a complementary metal–oxide–semiconductor-compatible fabrication progress.