
Lithography-free photonic processor for AI
Researchers in the US have created a photonic processor that provides programmable on-chip AI information processing without needing lithography.
The device developed at the University of Pennsylvania School of Engineering and Applied Science consists of spatially distributed optical gain and loss elements. Lasers cast light directly on a semiconductor wafer, without the need for defined lithographic pathways.
This uses dynamic control of spatial-temporal modulations of the imaginary index on an active semiconductor wafer. This tunes the optical gain distributions to route and switch optical signals to create neural networks. The processor has been tested with in situ training of vowel recognition with high accuracy.
“Photonic chips intended for machine learning applications face the obstacles of an intricate fabrication process where lithographic patterning is fixed, limited in reprogrammability, subject to error or damage and expensive,” said Liang Feng, Professor in the Departments of Materials Science and Engineering (MSE) and Electrical Systems and Engineering (ESE). “By removing the need for lithography, our chip overcomes those obstacles and offers improved accuracy and ultimate reconfigurability given the elimination of all kinds of constraints from predefined features.”
Without lithography, these chips become adaptable data-processing powerhouses. Because patterns are not pre-defined and etched in, the device is intrinsically free of defects. The lack of lithography also renders the microchip impressively reprogrammable, able to tailor its laser-cast patterns for optimal performance, whether for a simple task with few inputs and a small datasets or a complex task with many inputs and large datasets.
“What we have here is something incredibly simple,” says researcher Tianwei Wu. “We can build and use it very quickly. We can integrate it easily with classical electronics. And we can reprogram it, changing the laser patterns on the fly to achieve real-time reconfigurable computing for on-chip training of an AI network.”
“The interesting part,” says colleague Marco Menarini, “is how we are controlling the light. Conventional photonic chips are technologies based on passive material, meaning its material scatters light, bouncing it back and forth. Our material is active. The beam of pumping light modifies the material such that when the signal beam arrives, it can release energy and increase the amplitude of signals.”
“This active nature is the key to this science, and the solution required to achieve our lithography-free technology,” adds Zihe Gao who was also part of the team. “We can use it to reroute optical signals and program optical information processing on-chip.”
A lithography-free reconfigurable integrated photonic processor: https://www.nature.com/articles/s41566-023-01205-0