We propose and experimentally demonstrate an apodized bidirectional grating coupler for high-efficiency, perfectly vertical coupling. Through grating apodization, the coupling efficiency (CE) can be notably improved, and the parasitic reflections can be minimized. For ease of fabrication, subwavelength gratings are introduced, which are also beneficial for the coupling performance. Simulation shows a record CE of 72%. We found that the coupler is quite robust to the variation of incidence mode field diameter and fiber misalignment. A CE of $-1.8\mathrm{dB}$ is experimentally measured with a 1-dB bandwidth of 37 nm.

We report a vertically coupled germanium (Ge) waveguide detector integrated on silicon-on-insulator waveguides and an optimized device structure through the analysis of the optical field distribution and absorption efficiency of the device. The photodetector we designed is manufactured by IMEC, and the tests show that the device has good performance. This study theoretically and experimentally explains the structure of Ge PIN and the effect of the photodetector (PD) waveguide parameters on the performance of the device. Simulation and optimization of waveguide detectors with different structures are carried out. The device’s structure, quantum efficiency, spectral response, response current, changes with incident light strength, and dark current of PIN-type Ge waveguide detector are calculated. The test results show that approximately 90% of the light is absorbed by a Ge waveguide with 20 μm Ge length and 500 nm Ge thickness. The quantum efficiency of the PD can reach 90.63%. Under the reverse bias of 1 V, 2 V and 3 V, the detector’s average responsiveness in C-band reached 1.02 A/W, 1.09 A/W and 1.16 A/W and the response time is 200 ns. The dark current is only 3.7 nA at the reverse bias voltage of −1 V. The proposed silicon-based Ge PIN PD is beneficial to the integration of the detector array for photonic integrated arrayed waveguide grating (AWG)-based fiber Bragg grating (FBG) interrogators.

A four-port grating coupler with etched square holes is pro-posed and demonstrated for surface-normal and polarization independent coupling. Benefiting from the perfectly vertical coupling and 1×4 power splitting/combing scheme, efficient polarization independent operation was achieved. The cou-pling efficiency was further improved by grating apodization. According to the simulations, the proposed two-dimensional (2D) apodized grating coupler can achieve a coupling efficiency (CE) of 64.5% (-1.9dB), a low upward back-reflection of 8.5% (-11dB) and polarization-dependent loss (PDL) lower than 0.04dB across the wavelength range of 1520-1620nm. A CE of 56.3% (-2.5dB) was experimentally measured, with PDL be-low 0.4dB within the C-band and lower than 0.7dB within the L-band. Waveguide delay lines were used to compensate the waveguide channel phase differences at the output port.

In this work, a bidirectional grating coupler for perfectly vertical coupling is proposed. The coupling efficiency is enhanced using a silicon nitride (Si₃N₄) layer above a uniform grating. In the presence of Si₃N₄ layer, the back-reflected optical power into the fiber is diminished and coupling into the waveguide is increased. Genetic algorithm (GA) is used to optimize the grating and Si₃N₄ layer simultaneously. The optimal design obtained from GA shows that the average in-plane coupling efficiency is enhanced from about 57.5% (−2.5 dB) to 68.5% (−1.65 dB), meanwhile the average back-reflection in the C band is reduced from 17.6% (−7.5 dB) to 7.4% (−11.3 dB). With the help of a backside metal mirror, the average coupling efficiency and peak coupling efficiency are further increased to 87% (−0.6 dB) and 89.4% (−0.49 dB). The minimum feature size of the designed device is 266 nm, which makes our design easy to fabricate through 193 nm deep-UV lithography and lowers the fabrication cost. In addition, the coupler proposed here shows a wide-band character with a 1-dB bandwidth of 64 nm and 3-dB bandwidth of 96 nm. Such a grating coupler design can provide an efficient and cost-effective solution for vertical fiber-to-chip optical coupling of a Wavelength Division Multiplexing (WDM) application.

Optical receivers with potentially high operation bandwidth and low cost have received considerable interest due to rapidly growing data traffic and potential Tb/s optical interconnect requirements. Experimental realization of 65 GHz optical signal detection and 262 GHz intrinsic operation speed reveals the significance role of graphene photodetectors (PDs) in optical interconnect domains. In this work, a novel complementary metal oxide semiconductor post‐backend process has been developed for integrating graphene PDs onto silicon integrated circuit chips. A prototype monolithic optoelectronic integrated optical receiver has been successfully demonstrated for the first time. Moreover, this is a firstly reported broadband optical receiver benefiting from natural broadband light absorption features of graphene material. This work is a perfect exhibition of the concept of monolithic optoelectronic integration and will pave way to monolithically integrated graphene optoelectronic devices with silicon ICs for three‐dimensional optoelectronic integrated circuit chips.

Because of the von Neumann bottleneck, neuromorphic networks aimed at in-memory computing, such as brains, are extensively studied. As artificial synapses are essential in neuromorphic networks, a photonic synapse based on slot-ridge waveguides with nonvolatile phase-change materials (PCMs) was proposed and demonstrated in an SOI platform with standard complementary metal-oxide-semiconductor (CMOS) process for a larger weight dynamic range. The change of the optical transmission spectrum of our photonic synapses was about 3.5dB higher than that of primitive synapses, which meant large weight dynamic range and more weight values. A 90.7% recognition accuracy based on our photonic synapses, which was 2.6% higher than that of primitive synapses, was realized for the MNIST handwritten digits recognition task performed by a three-layer perceptron. Besides, because of the nonvolatile nature of PCMs, the weights achieved by our photonic synapses can be stored in situ ensuring a lower consumption in in-memory computing. This framework can potentially achieve a more efficient in-memory computing neuromorphic network in silicon photonics.