Silicon photonics is a field of study and technology that involves the use of silicon-based materials to generate, manipulate and detect light (photons) for various applications. Silicon is employed as a medium to harness the optical properties of light, and often involves integrating optical components with electronic components on the same silicon chip. Silicon photonic integrated circuits (PICs) are devices that combine both photonic and electronic components on a single silicon chip. The critical factors contributing to the widespread acceptance and economic viability of silicon PICs include processes that are compatible with CMOS and packaging techniques.
Building blocks of a silicon PIC
Active and passive devices are the fundamental elements of a silicon PIC. The use of passive devices negates the need for an external power supply. Discrete passive devices include waveguides, couplers and polarization splitters. Silicon PICs also rely on active devices in addition to passive ones. While passive devices, such as waveguides and couplers, manipulate light without external power, active devices introduce an element of control by actively modulating or amplifying the optical signals. These systems rely heavily on active devices like lasers, modulators and detectors made from silicon. Each silicon laser sends its output signal to the modulator via waveguides. After the silicon modulator encodes a signal into light, it travels over further waveguides before arriving at the silicon photodetectors, which transform the optical signals into electronic signals for further processing.
Waveguides
Silicon photonic circuits rely heavily on waveguides as a key passive component. In the 1980s, scientists claimed developing the first silicon waveguide. In silicon waveguides, thin strips or structures of silicon are used to confine and guide light along a specific path. These waveguides can be integrated into photonic devices and circuits on a silicon chip, allowing for the integration of both photonic and electronic components on the same platform. The compatibility of silicon waveguides with existing semiconductor technology makes them a key component in the development of integrated and efficient photonic circuits.
Couplers
An important challenge in creating commercial microphotonics devices is achieving efficient fiber-to-waveguide coupling. The mode mismatch between the silicon waveguides and the input/output fiber is large since the waveguides are submicron in size. An inverted taper coated in polymer or silicon oxide can be an effective strategy. The inverted taper broadens the optical mode to fit the modal size of the fiber, delocalizing the core mode profile. However, it has the limitation of being used only on the chip's periphery. A grating coupler, in which the periodic structure aids light coupling between the fiber and the waveguide, is an option. A grating coupler flips the direction of light travelling through a silicon waveguide to be perpendicular to the chip's surface. A grating coupler eliminates the need for facet polishing and effectively reduces back-reflections since light may be injected or withdrawn from any point on a chip.
Polarization splitter
Silicon PICs benefit greatly from careful polarization control. To get rid of the polarization sensitivity in silicon photonic nanowire waveguides, a polarization-diversity approach can be used in addition to the construction of polarization-independent waveguide components. By utilizing a polarization beam splitter (PBS) and a polarization rotator (PR), a polarization-diversity circuit may convert any input polarization into its orthogonal forms, transverse electrical (TE) and transverse magnetic (TM) polarizations. Two identical PICs then process the two distinct parts. One of the parts is then turned 90° by a second PR. The final step is the combination of the two orthogonal components by the second PBS. The PBS and PR are the two most crucial parts of this circuit.
Lasers
Semiconductor lasers predominantly use Group–V compounds like gallium arsenide, indium phosphide and gallium nitride, as opposed to silicon or germanium, due to their indirect bandgap nature. In materials like silicon or germanium, which have an indirect bandgap, free electrons typically reside in the lower valley of the conduction band, and this valley is not directly aligned with the free holes in the valence band. The indirect bandgap structure of silicon leads to a tendency for free electrons to recombine with holes, resulting in the release of phonons (heat) rather than photons. This characteristic gives rise to an exceptionally low internal quantum efficiency for light emission in silicon, estimated at around one photon per million electrons. In these materials, stimulated emissions often occur through processes mediated by phonons, where electrons absorb phonons to fulfill momentum conservation requirements and emit photons. The probability of such a phonon-mediated process is considerably lower compared to a straightforward recombination in materials with a direct bandgap.
Modulators and switches
Silicon PICs also incorporate crucial elements like silicon modulators and switches. A modulator is a device with the capability to change one or more characteristics of a transmitted light beam, particularly an optical carrier. On the flip side, a silicon switch enables the selective transition of an optical carrier within a silicon PIC from one circuit to another. Modulation/switching speed signifies the maximum data rate that can be applied to an optical carrier before the modulation/switching amplitude diminishes to 3 dB. Optical carriers can undergo modulation or switching through the utilization of either non-optical means (heat, electrical field) or optical signals.
Photodetectors
Photodetectors made of silicon, positioned at the optical interconnect’s end in a silicon PIC, convert optical signals into electronic signals. Operating through the absorption of incident photons, these photodetectors generate pairs of free electrons and holes, which are then guided to the electrodes by an electric field. This process yields an electric current, referred to as photocurrent. To generate electron-hole pairs, the energy of incident photons must be equal to or greater than the bandgap energy.
Conclusion
Silicon PICs are devices that combine both photonic and electronic components on a single silicon chip. These circuits leverage the optical properties of silicon to manipulate and transmit light signals, enabling various functionalities such as data transmission, processing and sensing. Silicon PICs typically include passive devices like waveguides, couplers and interferometers for guiding and manipulating light, as well as active devices like modulators and detectors for modulating and detecting optical signals.
