Search for: All records

Abstract With success of silicon photonics having mature to foundry-readiness, the intrinsic limitations of the weak electro-optic effects in Silicon limit further device development. To overcome this, heterogeneous integration of emerging electrooptic materials into Si or SiN platforms are a promising path to deliver <1fJ/bit device-level efficiency, 50+Ghz fast switching, and <10's um^2 compact footprints. Graphene's Pauli blocking enables intriguing opportunities for device performance to include broadband absorption, unity-strong index modulation, low contact resistance. Similarly, ITO has shown ENZ behavior, and tunability for EOMs or EAMs. Here we review recent modulator advances all heterogeneously integrated on Si or SiN such as a) a DBR-enabled photonic 60 GHz graphene EAM, b) a hybrid plasmon graphene EAM of 100aJ/bit efficiency, d) the first ITO- based MZI showing a VpL = 0.52 V-mm, and e) a plasmonic ITO MZI with a record low VpL = 11 V- um. We conclude by discussing modulator scaling laws for a roadmap to achieve 10's aJ/bit devices. 

Abstract Photonic neural networks (PNN) are a promising alternative to electronic GPUs to perform machine-learning tasks. The PNNs value proposition originates from i) near-zero energy consumption for vector matrix multiplication once trained, ii) 10-100 ps short interconnect delays, iii) weak required optical nonlinearity to be provided via fJ/bit efficient emerging electrooptic devices. Furthermore, photonic integrated circuits (PIC) offer high data bandwidth at low latency, with competitive footprints and synergies to microelectronics architectures such as foundry access. This talk discusses recent advances in photonic neuromorphic networks and provides a vision for photonic information processors. Details include, 1) a comparison of compute performance technologies with respect to compute efficiency (i.e. MAC/J) and compute speed (i.e. MAC/s), 2) a discussion of photonic neurons, i.e. perceptrons, 3) architectural network implementations, 4) a broadcast-and-weight protocol, 5) nonlinear activation functions provided via electro-optic modulation, and 6) experimental demonstrations of early-stage prototypes. The talk will open up answering why neural networks are of interest, and concludes with an application regime of PNN processors which reside in deep-learning, nonlinear optimization, and real-time processing.