An optical isolator allows light waves to propagate in a specified direction and not in the opposite. By virtue of this behavior, the isolator plays an essential role in preventing undesired optical feedback from interacting with optical active devices. In this lecture, various types of optical isolators that are integratable on silicon phonic platforms are presented including non-magneto-optical devices.

Most part of lecture focuses on a magneto-optical isolators fabricated in silicon waveguides. The lecture addresses the device design principles as well as the state-of-the-art performance characteristics demonstrated in fabricated devices.

The lecture provides a panorama of the potentials and limits of the silicon photonic platform to implement passive functionalities. Aspects apparently trivial or negligible that can have a large impact on the overall performance of the entire circuit will be considered with detail. The state of the art of passive devices will be shown and rings resonators will be treated in detail, starting from an historical survey and going through theory and applications. Several arguments have the scope to trigger further interests and are propaedeutic to the other lectures.

Since silicon nanophotonic waveguides usually have ultra-high birefringence due to the ultra-high index-contrast and the nano-scale cross section, polarization management is playing an important role in silicon photonics. In this lecture, I will discuss new technologies and silicon photonic devices for on-chip polarization management, including polarizers, polarization beam splitters, as well as polarization rotators. The challenges in this field will also be discussed.

Standard silicon phase modulators based on carrier dispersion effects suffer from a relatively low efficiency and/or high loss, and residual amplitude modulation.  Therefore several alternatives are being investigated. Non-standard materials such as BTO, Lithium Niobate, PZT, EO-polymers or 2D-materials are integrated using heterogeneous integration methods and do allow for pure phase modulation.  We will review state-of-the-art, open challenges and end with prospects for future work.

The analysis of noise is an essential aspect of optical communications and sensing, but also one of the more subtle ones. I will cover the physical origin and modeling of laser, optical amplifier and transceiver noise with the objective to enable a comprehensive point-to-point system analysis. Application examples in both optical communications and metrology will be given.

In this lecture I will discuss micro-transfer printing as a novel technique to realise heterogeneous silicon photonic integrated circuits, through the integration of III-V opto-electronic components and other devices/materials on the silicon photonics platform.

We will introduce the field of large-scale programmable photonics. In the past few years, new concepts for general-purpose, programmable photonic integrated circuits (PIC) have been proposed to manipulate light in a more flexible way. Today, most PICs are developed for one function, and are therefore called application-specific photonic integrated circuits (ASPIC), and a new chip has to be designed for each new application The new class of programmable PICs can be used more like electronic microcontrollers and field-programmable gate arrays (FPGA), and programmed in software for different tasks. Simple programmable PICs have already been experimentally demonstrated, reproducing functionality in programming hitherto limited to custom-designed hardware. This programming is done through electro-optic tuners like waveguide heaters or MEMS. We will discuss different architectures for programmable PICs, and the technology stack that is needed to realize the full programmable PICs: waveguides, actuators, monitors, control loops, programming algorithms and a user interface that makes the hardware accessible to the users. We will briefly discuss the economics behind programmable PICs, and draw parallels with programmable electronics, which have put complex functionality in the hands of the maker community.

Photonic device packaging can account for over 50% of the photonic product manufacturing cost. Therefore, industries developing photonic-based products must understand the materials, technologies and processes required to package photonic devices. In this short lecture, we will review some of the main principles of photonic packaging and how they apply to Silicon Photonics. This includes an overview of some key packaging and assembly building blocks, with examples of how they are applied in real-world examples.

Machine learning has made tremendous progress recently, as evidenced by the success of deep learning and neuromorphic photonics. In this lecture, we will discuss how silicon photonics can be an interesting hardware platform for the implementation of these paradigms. We will focus mostly on a technique called reservoir computing, and illustrate how it can be used to perform e.g. non-linear dispersion compensation in telecom links, or to identify different kinds of particles in a flow cytometer setup.

By allowing for arbitrary device geometries, inverse design methods enable nanophotonics devices with improved efficiencies, reduced footprints, and novel functionalities. We explore how these new design methods can be leveraged to improve integrated photonics further and enable new applications.