Complex-amplitude metasurface-based orbital angular momentum holography in momentum space

Optical tweezers use the forces exerted by a strongly focused beam of light to trap and move objects ranging in size from tens of nanometres to tens of micrometres. Since their introduction in 1986, the optical tweezer has become an important tool for research in the fields of biology, physical chemistry and soft condensed matter physics. Recent advances promise to take optical tweezers out of the laboratory and into the mainstream of manufacturing and diagnostics; they may even become consumer products. The next generation of single-beam optical traps offers revolutionary new opportunities for fundamental and applied research.

Entangled quantum states are not separable, regardless of the spatial separation of their components. This is a manifestation of an aspect of quantum mechanics known as quantum non-locality. An important consequence of this is that the measurement of the state of one particle in a two-particle entangled state defines the state of the second particle instantaneously, whereas neither particle possesses its own well-defined state before the measurement. Experimental realizations of entanglement have hitherto been restricted to two-state quantum systems, involving, for example, the two orthogonal polarization states of photons. Here we demonstrate entanglement involving the spatial modes of the electromagnetic field carrying orbital angular momentum. As these modes can be used to define an infinitely dimensional discrete Hilbert space, this approach provides a practical route to entanglement that involves many orthogonal quantum states, rather than just two Multi-dimensional entangled states could be of considerable importance in the field of quantum information, enabling, for example, more efficient use of communication channels in quantum cryptography.

Deep learning has been transforming our ability to execute advanced inference tasks using computers. We introduce a physical mechanism to perform machine learning by demonstrating an all-optical Diffractive Deep Neural Network (D2NN) architecture that can implement various functions following the deep learning-based design of passive diffractive layers that work collectively. We create 3D-printed D2NNs that implement classification of images of handwritten digits and fashion products as well as the function of an imaging lens at terahertz spectrum. Our all-optical deep learning framework can perform, at the speed of light, various complex functions that computer-based neural networks can implement, and will find applications in all-optical image analysis, feature detection and object classification, also enabling new camera designs and optical components that perform unique tasks using D2NNs.