Ways to Put Lasers on Silicon
spectrum.ieee.org
spectrum.ieee.org
I have a strange reason for asking. My wife did most of the illustrations for this article, but not that one. It was sent along as part of the information packet for her to work from, but we had trouble figuring out what it was actually showing. She asked the art director, but I wasn't sure if he was answering the right question. She eventually did a version of it, but they didn't put it in, at least yet. So I thought maybe someone here would know!
The bright cubes are the laser chips that are bonded on top of the quartz, since native silicon lasers aren't practical.
That whole illustration looks like a bunch of chips in a wafer maybe with testing and monitoring patterns included.
Looks like some other stuff like modulators in the top right of the square. Maybe a test run for process control?
Edit: yeah the cubes are definitely laser dies, fig 2 shows the process. The patterns are the couplers and such to get the light in without losses and at the right direction.
In some ways it's fascinating that illustrations for technical articles work at all. Usually there are two layers of people who may not understand the tech between the artist and the scientists, and no direct communication allowed. Choices often come down to Googling to find earlier publications online by the same group, and hoping those earlier illustrations were accurate.
The idea is simply place a laser die and the passive die next to each other during packaging, and the laser power is transferred between the dies by butt-coupling (or photonic wire bonding). The R&D cost can be much lower but the tedious (and usually required to be active) alignment procedure prevents mass-production.