For roughly two decades, photonic crystals—uniform lattice patterns repeated with unwavering precision—have served as the fundamental design principle for controlling light inside semiconductor lasers. Only when diffraction gratings are arranged with strict regularity can light resonate at a specific wavelength and emerge from the substrate, perpendicular to its surface, as a single, phase-coherent beam. While this geometric rigor has been the greatest strength underpinning the quality of laser light, it has also imposed manufacturing constraints—since the fine air-hole structures are prone to collapse during the thermal processes of crystal growth—and design limitations, since resonators with differing characteristics cannot be fabricated side-by-side on the same wafer.
In 2026, a research group led by Professor Kent D. Choquette of the Department of Electrical and Computer Engineering at the University of Illinois Urbana-Champaign (UIUC), together with doctoral student Erin M. Raftery, overcame this premise in the laboratory. The team formed a quasi-periodic pattern—one containing only partial periodicity rather than strict regularity—inside a semiconductor structure, and succeeded in generating infrared laser emission at a wavelength of 1.5 at room temperature. The findings were published in the journal Applied Physics Letters (DOI: 10.1063/5.0325678).
This is a laboratory-stage demonstration using optical pumping, in which an external light source provides the excitation energy—it does not immediately translate into a practical laser diode. Nevertheless, the finding that an intentionally disordered array can still function stably as a resonator for a surface-emitting laser carries significant weight.
The Manufacturing Challenge: Fine Air Holes That Collapse Under Heat in Conventional Surface-Emitting Lasers
Conventional photonic-crystal surface-emitting lasers (PCSELs) operate by introducing a two-dimensional periodic variation in refractive index inside a semiconductor. Light generated in the active layer spreads across a wide in-plane area through a lattice-shaped array of structures and interferes with itself through diffraction. Only light of a specific wavelength constructively interferes, and it is emitted perpendicular to the device surface with aligned phase. This structure yields an extremely narrow beam divergence angle and a nearly circular cross-section, producing high-quality light compared with conventional edge-emitting lasers. PCSELs have been actively researched for roughly 20 years for advanced applications in aerospace, defense systems, and optical communications.
However, the excellent optical characteristics of PCSELs rest on extremely strict manufacturing conditions. In a typical PCSEL, a semiconductor layer is partially etched to form an array of nanoscale air holes, and the refractive-index difference between the semiconductor and air is used to confine light. A further semiconductor layer must then be grown (regrown) on top of this air-hole layer to bury it.
Here a physical problem arises. Under the high-temperature conditions of crystal growth, a phenomenon called mass transport occurs, in which semiconductor atoms migrate across the surface, causing air holes that were originally fabricated to a precise design to deform, or causing unintended merging of adjacent cavities. This distortion associated with regrowth is especially pronounced in indium phosphide (InP)-based material systems, which are widely used in optical communications. Even a slight deviation in the shape or spacing of the air holes disrupts the resonance conditions for light, undermining the uniformity and structural integrity of the laser.
Any attempt to design more complex patterns—varying the shape or arrangement of air holes region by region—causes manufacturing yield to drop precipitously. As a result, PCSEL design has remained strongly bound by the geometric constraint of repeating a single, identical fine pattern uniformly across the entire substrate.
Eliminating Air Holes in Favor of an Embedded Dielectric, and Breaking the Periodicity of the Array
To overcome this manufacturing challenge, Professor Choquette's research group has been exploring an approach that dispenses with air holes altogether. As a precursor to this work, the team published foundational technology in 2025 in the IEEE Photonics Journal (DOI: 10.1109/JPHOT.2025.3561087), in which a fine pattern of silicon dioxide () is embedded inside a semiconductor in place of air holes. Rather than digging vertically into the semiconductor layer and leaving a cavity, the structure involves regrowing an epitaxial semiconductor layer over an etched pattern layer to cover it.
Because a low-refractive-index solid material— at sub-micron dimensions—is embedded within the high-refractive-index semiconductor, the shape resists deformation even through the high-temperature regrowth process. Compared with a method that leaves air-filled cavities, structural strength is preserved, enabling more reproducible manufacturing.
With the conventional air-hole approach, a semiconductor layer is etched to form cavities before the semiconductor is regrown, meaning the risk of cavity deformation under high heat is a constant concern. By contrast, in the embedded-dielectric approach used in this research, the solid material is pre-positioned and embedded between semiconductor layers, making it physically easier to retain the designed shape even through thermal processing.
In the present study, the team pushed the capabilities of this embedded-dielectric platform further. Departing from the fully periodic lattice structure that conventional PCSELs have long adhered to, they extended the design to a quasi-periodic pattern containing only partial periodicity.
The design was inspired by recent research in condensed matter physics on topologically protected aperiodic structures. However, the laser device itself in this study does not exhibit the physical properties of topological protection. Rather than confining the arrangement to strictly equal spacing and identical shapes, the research group designed a quasi-photonic crystal structure containing partial periodicity, in which the array varies according to precisely controlled rules.
Regarding the intent behind this design, co-author Erin M. Raftery stated, "We achieved a structure without a periodic pattern that can be tuned with much more flexibility. It's a different engineering approach to modulating the refractive index in order to draw out the characteristics we want from the laser." It is precisely because the embedded dielectric suppresses thermal deformation during manufacturing that such a complex, asymmetric pattern can be fixed inside the device without collapsing.
| Device Type | Low-Index Region | Pattern Periodicity | Demonstrated Operation | Key Limitations | Development Status |
|---|---|---|---|---|---|
| Conventional PCSEL | Air holes (cavities) | Fully periodic structure | Current injection or optical pumping | Risk of air-hole deformation during regrowth | Practical use, partially commercialized |
| 2025 Precursor Device | Silicon dioxide () | Fully periodic structure | Optical pumping (pulsed light) | Practical current injection not yet achieved | Proof-of-concept for foundational technology |
| This Study (2026 QPCSEL) | Silicon dioxide () | Quasi-periodic structure with partial periodicity | Optical pumping | Optical pumping only; device performance not yet evaluated | Demonstration of quasi-periodic structure operation |
Room-Temperature Lasing at 1.5 Micrometers and the Boundaries of the Measured Data
In evaluating the completed quasi-periodic photonic-crystal surface-emitting laser (QPCSEL), the research team confirmed laser emission at a wavelength of 1.5 under room-temperature conditions. In the paper, the authors report that they "demonstrated optically pumped lasing at room temperature with an emission wavelength of 1.5 from the embedded-dielectric QPCSEL."
The 1.5 wavelength band coincides with the standard band for optical communications, where transmission loss in silica optical fiber is lowest. However, this coincidence stems from the choice of an indium-phosphide-based material system suited to long-distance optical communications; no actual optical-communication transmission test was conducted in this study.
Some technology media reports have described the result as confirming "single-mode lasing," but the University of Illinois's official press materials and initial Phys.org coverage stop at stating the fact that the device "lased successfully at room temperature." Detailed quantitative characteristics of the laser—such as single-mode purity, threshold lasing power, side-mode suppression ratio, and beam quality—remain to be verified against the precise data presented in the original peer-reviewed paper.
Clarity regarding the measurement conditions is also needed. The 2025 precursor paper described a measurement method using pulsed 980 nm pump light with a spot diameter of approximately 200 , operating the device without active cooling. However, the extent to which the same measurement parameters were applied to this quasi-periodic-structure device in the present study must be confirmed against the individual experimental setup; the figures from the precursor device cannot simply be carried over.
Furthermore, this report does not include direct performance comparison data against a standard periodic PCSEL fabricated under identical conditions, nor does it include manufacturing yield or the range of statistical variation for the device. Whether the quasi-periodic structure offers advantages over the periodic structure in terms of laser output power or power-conversion efficiency remains, at this point, unverified.
The Structural Gap Between Proof of Concept and a Practical Device
The significance of this achievement lies in demonstrating the "physical feasibility" that a quasi-periodic structure containing partial periodicity can indeed function as a surface-emitting laser. But a test chip fabricated in a laboratory does not immediately translate into an industrial laser light source.
The greatest barrier lies in the device's driving method. The device in this study was operated via "optical pumping," in which energy is supplied by irradiating it with a separate external laser beam. To incorporate it into electronic circuitry as a semiconductor chip and mount it in communication equipment or sensors, the structure must be advanced to a "current-injection" diode laser, in which current is passed directly through electrodes to induce lasing.
In a structure containing an embedded dielectric layer, designing carrier transport to flow current uniformly and efficiently around the electrically insulating and inject it into the active layer becomes extremely difficult. Professor Choquette himself has acknowledged that this represents a more challenging next step, stating, "We've demonstrated the physics. Now we need to demonstrate a practical device."
Moreover, the hypothesis that the embedded dielectric improves reliability remains, for now, a theoretical expectation. The idea that filling the structure with —which has higher thermal conductivity than air—improves heat dissipation during operation and prevents atomic diffusion over the device's lifetime was presented as a possibility in the 2025 paper, but it has not been proven through long-term operational testing in this study.
The Engineering Possibilities Opened Up by Integration on a Single Substrate
While challenges toward practical implementation remain substantial, the design freedom afforded by the combination of a quasi-periodic structure and an embedded dielectric offers a new perspective on methods for manufacturing photonic integrated circuits.
In conventional PCSEL manufacturing, the basic approach has been to grow a single periodic pattern uniformly across the entire wafer. Fabricating multiple lasers with different wavelengths or different beam shapes side by side on the same substrate has been considered extremely difficult, since each would carry a different risk of deformation during regrowth. Professor Choquette notes, "Right now, we can only grow one type of structure at a time, but using this method, it becomes possible to combine different structures on the same substrate."
If this flexibility is established, it may become possible to integrate on a single semiconductor chip multiple laser light sources with slightly different wavelengths, or light sources with different beam profiles for wide-angle illumination versus focused beams. The research group and related industries have cited LiDAR (a distance-measurement technology using light), remote sensing, optical navigation, ultra-compact optical communication modules, and aerospace equipment as potential future applications.
These remain future application prospects made possible by an extension of design methodology; the present study does not verify implementation in any specific system. No commercialization roadmap, licensing arrangements, or independent replication by other institutions have been reported at this time.
What the research team has demonstrated is an engineering foothold for removing the reliance on "fully periodic structures" that has dominated semiconductor laser resonator design for roughly two decades. What optical output and electrical efficiency can a photonic crystal—freed from geometric constraints through the manufacturing innovation of an embedded dielectric—achieve in a future current-injection device? The true optical utility offered by irregular patterns containing partial periodicity will be put to the test by future advances in device engineering.