High-efficiency self-focusing metamaterial grating coupler in silicon nitride with amorphous silicon overlay.

Efficient fiber-chip coupling interfaces are critically important for integrated photonics. Since surface gratings diffract optical signals vertically out of the chip, these couplers can be placed anywhere in the circuit allowing for wafer-scale testing. While state-of-the-art grating couplers have been developed for silicon-on-insulator (SOI) waveguides, the moderate index contrast of silicon nitride (SiN) presents an outstanding challenge for implementing efficient surface grating couplers on this platform.

Low-loss grating coupler based on inter-layer mode interference in a hybrid silicon nitride platform.

Surface grating couplers are an important component for interfacing photonic integrated circuits with optical fibers. However, conventional coupler designs typically provide limited performance due to low directionality and poor fiber-to-grating field overlap. The efficiency can be improved by using non-uniform grating structures at the expense of small critical dimensions complicating the fabrication process.

Experimentally realized in situ backpropagation for deep learning in photonic neural networks.

Integrated photonic neural networks provide a promising platform for energy-efficient, high-throughput machine learning with extensive scientific and commercial applications. Photonic neural networks efficiently transform optically encoded inputs using Mach-Zehnder interferometer mesh networks interleaved with nonlinearities. We experimentally trained a three-layer, four-port silicon photonic neural network with programmable phase shifters and optical power monitoring to solve classification tasks using "in situ backpropagation," a photonic analog of the most popular method to train conventional neural networks.

Variable optical true-time delay line breaking bandwidth-delay constraints.

Continuously variable true-time optical delay lines are typically subject to a constraint of the bandwidth-delay product, limiting their use in several applications. In this Letter, we propose an integrated topology that breaks the bandwidth-delay product limit. The device is based on multiple Mach-Zehnder Interferometers (MZIs) arranged in parallel, providing easier control and a larger bandwidth compared to ring resonator-based solutions.

Power monitoring in a feedforward photonic network using two output detectors.

Programmable feedforward photonic meshes of Mach-Zehnder interferometers are computational optical circuits that have many classical and quantum computing applications including machine learning, sensing, and telecommunications. Such devices can form the basis of energy-efficient photonic neural networks, which solve complex tasks using photonics-accelerated matrix multiplication on a chip, and which may require calibration and training mechanisms. Such training can benefit from internal optical power monitoring and physical gradient measurement for optimizing controllable phase shifts to maximize some task merit function.

Separating arbitrary free-space beams with an integrated photonic processor.

Free-space optics naturally offers multiple-channel communications and sensing exploitable in many applications. The different optical beams will, however, generally be overlapping at the receiver, and, especially with atmospheric turbulence or other scattering or aberrations, the arriving beam shapes may not even be known in advance. We show that such beams can be still separated in the optical domain, and simultaneously detected with negligible cross-talk, even if they share the same wavelength and polarization, and even with unknown arriving beam shapes.

Dynamic mitigation of nonlinear effects in a silicon photonic add-drop filter.

Nonlinear effects limit the maximum amount of optical power that can be handled by silicon photonic integrated circuits (PICs). This limitation is particularly tight in resonant devices such as microring resonator (MRR) filters, suffering from a power-dependent resonance spread due to intracavity power enhancement. In this Letter, we present an automatic control system that can dynamically mitigate the nonlinear spectral distortion of silicon MRR filters by thermally controlling each MRR.

Polarization-transparent silicon photonic add-drop multiplexer with wideband hitless tuneability.

Flexible optical networks require reconfigurable devices with operation on a wavelength range of several tens of nanometers, hitless tuneability (i.e. transparency to other channels during reconfiguration), and polarization independence.

Electrical conductance of silicon photonic waveguides.

Many optoelectronic devices embedded in a silicon photonic chip, like photodetectors, modulators, and attenuators, rely on waveguide doping for their operation. However, the doping level of a waveguide is not always reflecting in an equal amount of free carriers available for conduction because of the charges and trap energy states inevitably present at the / interface. In a silicon-on-insulator technology with 10-doped native waveguides, this can lead to a complete depletion of the core from free carriers and to a consequently very high electrical resistance.