WHAT ARE OPTICAL CHIPS

As we mentioned in the introduction, optical chips are normal silicon-based chips that have some optical parts. The fact that we mention "optical" doesn't mean exactly that works with light as we know it, it means that it works with photons rather than electrons. Optical chips are usually referred as "Photonic Integrated Chips", or PICs, and the field of study of such PICs is often called "Photonics" or "Optoelectronics", which can be defined as "the technology of generating and harnessing light and other forms of radiant energy whose quantum unit is the photon". The range of applications of photonics extends from energy generation, modulation and detection to information storage, processing and communication. "Photonics incorporates the fields of optics, electrical engineering, physics, chemistry and materials science".

Since a chip is silicon-based and works with electrons, we need some kind of translation between electrons and photons and viceversa. Ms Victoria Steblina, from the Optical Sciences Center, in her document about optical circuits mentions that "One problem with optical circuitry is that it needs to be connected to optical fibers, a difficult process as the fibers and circuits have different geometries". She points out that optical fiber is usually circular while circuits are square "creating problems with the alignment and attachment of fibers to the edge of the circuits". The newest technology used to incorporate optical circuits consists on the "use of intense laser light to literally 'write' the optical part of the chip" (See New Technologies). Science Magazine, in the article "Fast Ferroelectric Liquid-Crystal Electrooptics" shows an optical chip that "can convert electronic data to optical form at about 0.5 Gbit/s".

How is the conversion done? One way is done by using Optical Detectors. The most common types of detectors for optical communications are PIN diodes and avalanche diodes.

Roy Blake, in the book "Basic Electronic Communication" explains how they work. He explains that in the PIN diode, "photons create electron-hole pairs in the depletion zone". If a photon has enough energy to move an electron from the valence to the conduction band, it can, by colliding with an electron, create a free electron. The PIN diode is reversed-biased, therefore the free electron and the hole will travel in opposite directions, creating a current that is proportional to the light power. Since the depletion zone is very wide, the PIN diode allows very high-speed operations. Roy Blake also shows how the avalanche photodiode works.

PIN-Diode Photodetector

The avalanche photodiode (APD), is also reverse-biased. The difference with the PIN diode is that the absorption of a photon of incoming light may set off an electron-hole pair avalanche breakdown, creating up to 100 more electron-hole pairs. "This feature gives the APD high sensitivity" (much greater than the PIN diode).

One phenomenon associated with this photon-electron conversion is the so called Dark current, which is the leakage current in the absence of light (is the normal reverse current of the diode). This dark current is due to electron-hole pairs created by thermal activity, and therefore it increases with temperature.

What about the other way around? How do you convert from electricity to light? If a photon can liberate an electron, something similar can be done but in the opposite way. The movement of an electron from a higher energy level to a lower energy level can cause the emission of a photon. In order to achieve that we can use Light-Emitting Diodes or Laser Diodes. "Basic Electronic Communication" shows how they work.

A light-Emitting diode (LED) is forward biased. The recombination of electron-hole pairs in any junction diode causes energy to be released. With an ordinary silicon diode, this energy is released in the form of heat, but in a LED, a significant proportion of the energy is radiated as photons of either visible or infrared light (IRED).

The Laser diode works similarly to the LED. The difference is that "this diode has a junction surrounded by material that has a lower refractive index than the material used for the junction". The ends of the junction have partly transmissive mirrors " so light can escape in one direction only. The whole junction area forms a resonant cavity at the operating frequency". Laser diodes operate with high current densities, which causes many electron-hole pairs to be generated. The recombination of these pairs generates photons which stimulate other pairs to combine. Roy Blake explains that "Laser operation, sometimes called lasing, occurs only above a certain threshold current". Below that current, the laser will behave in the same way a LED would.