Quasi‑Periodic Semiconductor Laser Breaks the Mold of Conventional Designs

When you think of a laser, you probably picture a perfectly ordered crystal lattice guiding photons in a neat, predictable way. That’s the textbook picture for photonic‑crystal surface‑emitting lasers (PCSELs), which rely on a regular, repeating pattern etched into a semiconductor.

But a research group at the University of Illinois Urbana‑Champaign decided to toss that rulebook aside. Instead of a strict lattice, they introduced a quasi‑periodic arrangement – a pattern that changes its spacing in a controlled, but non‑repeating fashion. The result? A laser that still spits out a clean, single‑mode beam at the telecom‑friendly wavelength of 1.5 µm, and it does so at room temperature.

The trick lies in a buried‑dielectric platform the team had pioneered earlier. Rather than carving holes directly into the active semiconductor, they first laid down a thin layer of silicon‑dioxide, patterned it with the desired quasi‑periodic geometry, and then grew the semiconductor on top. Those dielectric features are safely hidden beneath the epitaxial layer, so they survive the harsh regrowth process without distortion.

In the lab, the device was photopumped – an external laser supplied the energy – and it showed unmistakable single‑mode lasing. “We’ve proved that you don’t need a perfectly periodic crystal to achieve tight refractive‑index control,” says Erin Raftery, a PhD candidate on the project. The quasi‑periodic crystal still creates the required feedback, but now designers have an extra degree of freedom to tweak the pattern for specific needs.

Why does this matter? Conventional PCSELs are powerful, but their strict geometry can make fabrication tricky, especially when you try to stack different designs on the same wafer. With the buried‑dielectric approach, you could, in principle, mix several distinct photonic‑crystal patterns on a single chip, tailoring each laser for tasks ranging from lidar in autonomous cars to high‑speed optical communication.

The work is still at the proof‑of‑concept stage. The laser is photopumped, not yet electrically injected, so it isn’t ready for commercial diode‑laser applications. Still, the physics is solid, and the team’s next milestone is an electrically driven version that could make its way into real‑world systems.

Published in Applied Physics Letters, this study opens a fresh design space for semiconductor lasers, suggesting that a little randomness—when carefully engineered—might be the key to more versatile, reliable light sources.