Photonic myths

Paul van Gerven is an editor at Bits&Chips.

Leestijd: 3 minuten

Bits&Chips once met a Dutch local politician who was under the impression that integrated photonics would replace all electronics. Surely, photonics lobbyists didn’t intentionally impart that impression on him, but the sector should take the existence of such misunderstandings to heart nonetheless. Setting expectations so unrealistically high that they can never be met will eventually result in a backlash, endangering the entire effort to build an integrated photonics industry in the Netherlands. Not to mention the fact that it – understandably – pisses off the electronics folks to no end.

One day, an ‘optical transistor’ may be invented that can compete with the electron-based version, but current photonic integrated circuit (PIC) technology will never supersede CMOS. A truly optical computing element cannot be smaller than the wavelength of the light passing through it, which is at least several hundreds of nanometers. That’s huge compared to advanced CMOS, whose smallest features measure tens of nanometers. Thus, PICs can never reach the same complexity of ICs – the optical ICs would get gigantic.

There are PIC applications, of course, that don’t require IC-like complexity. The primary example is signal routing in fiber-optic communication networks. This is currently performed by optical-electronic-optical conversion, which limits data rates. All-optical signal processing would greatly speed things up. With demand for data transport ever-increasing, this is one area where PICs can really shine.

This article is exclusively available to premium members of Bits&Chips. Already a premium member? Please log in. Not yet a premium member? Become one and enjoy all the benefits.


Related content