Data transmission rates in optical communication systems are approaching the limits of conventional multiplexing methods. Orbital angular momentum (OAM) in optical vortex beams offers a new degree of freedom and the potential to increase the capacity of free-space optical communication systems, with OAM beams acting as information carriers for OAM division multiplexing (OAM-DM). We demonstrate independent collinear OAM channel generation, transmission and simultaneous detection using Dammann optical vortex gratings (DOVGs). We achieve 80/160 Tbit s⁻¹ capacity with uniform power distributions along all channels, with 1600 individually modulated quadrature phase-shift keying (QPSK)/16-QAM data channels multiplexed by 10 OAM states, 80 wavelengths and two polarizations. DOVG-enabled OAM multiplexing technology removes the bottleneck of massive OAM state parallel detection and offers an opportunity to raise optical communication systems capacity to Pbit s⁻¹ level.

Scattering forces in focused light beams push away metallic particles. Thus, trapping metallic particles with conventional optical tweezers, especially those of Mie particle size, is difficult. Here we investigate a mechanism by which metallic particles are attracted and trapped by plasmonic tweezers when surface plasmons are excited and focused by a radially polarized beam in a high-numerical-aperture microscopic configuration. This contrasts the repulsion exerted in optical tweezers with the same configuration. We believe that different types of forces exerted on particles are responsible for this contrary trapping behaviour. Further, trapping with plasmonic tweezers is found not to be due to a gradient force balancing an opposing scattering force but results from the sum of both gradient and scattering forces acting in the same direction established by the strong coupling between the metallic particle and the highly focused plasmonic field. Theoretical analysis and simulations yield good agreement with experimental results.

On-chip twisted light emitters are essential components of orbital angular momentum (OAM) communication devices^{1, 2}. These devices address the growing demand for high-capacity communication systems by providing an additional degree of freedom for wavelength/frequency division multiplexing (WDM/FDM). Although whispering-gallery-mode-enabled OAM emitters have been shown to possess some advantages^{3, 4, 5}, such as compactness and phase accuracy, their inherent narrow bandwidths prevent them from being compatible with WDM/FDM techniques. Here, we demonstrate an ultra-broadband multiplexed OAM emitter that utilizes a novel joint path-resonance phase control concept. The emitter has a micron-sized radius and nanometer-sized features. Coaxial OAM beams are emitted across the entire telecommunication band from 1,450 to 1,650 nm. We applied the emitter to an OAM communication with a data rate of 1.2 Tbit/s assisted by 30-channel optical frequency combs (OFCs). The emitter provides a new solution to further increase capacity in the OFC communication scenario.

An optical diffractive neural network (DNN) can be implemented with a cascaded phase mask architecture. Like an optical computer, the system can perform machine learning tasks such as number digit recognition in an all-optical manner. However, the system can work only under coherent light illumination, and the precision requirement in practical experiments is quite high. This Letter proposes an optical machine learning framework based on single-pixel imaging (MLSPI). The MLSPI system can perform the same linear pattern recognition task as DNN. Furthermore, it can work under incoherent lighting conditions, has lower experimental complexity, and can be easily programmable.