🔬 optics

Low-Threshold Surface-Emitting Whispering-Gallery Mode Microlasers

This contribution reports on the realization of low-threshold, surface-emitting micro-lasers in the Whispering-Gallery mode using smooth AlGaAs micropillars with high-quality distributed Bragg reflectors, achieving simultaneous comb-like laser action in the 930–970 nm range and a transition to single-mode operation at 130 K for 5 μm pillars with an estimated threshold of 240 μW.

Original authors: Andrey Babichev, Ivan Makhov, Natalia Kryzhanovskaya, Sergey Troshkov, Yuriy Zadiranov, Yulia Salii, Marina Kulagina, Mikhail Bobrov, Alexey Vasilev, Sergey Blokhin, Nikolay Maleev, Leonid Karachinsky

Published 2026-04-30

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Andrey Babichev, Ivan Makhov, Natalia Kryzhanovskaya, Sergey Troshkov, Yuriy Zadiranov, Yulia Salii, Marina Kulagina, Mikhail Bobrov, Alexey Vasilev, Sergey Blokhin, Nikolay Maleev, Leonid Karachinsky, Innokenty Novikov, Anton Egorov

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Idea: Tiny Light Traps

Imagine you have a tiny, perfect cylinder made of glass (in this case, a semiconductor pillar about as wide as a human hair). Inside this cylinder are special "light traps" called Whispering-Gallery Modes (WGM).

Think of a Whispering Gallery like the dome of St. Paul's Cathedral in London. If you whisper against the curved wall on one side, the sound travels all the way around the curve to the other side without fading. In these tiny pillars, light does the same thing: it races around the inner edge of the cylinder, bouncing off the walls instead of shooting straight up and down.

The scientists in this paper wanted these light traps to function as lasers (intense, focused beams of light) that shoot straight out from the top of the pillar, rather than exiting sideways.

The Problem: The "Leaky" Roof

Normally, scientists use mirrors at the top and bottom to keep light trapped inside these tiny cylinders. However, the mirrors they previously used were like "leaky roofs." They absorbed too much of the energy that entered, meaning the laser required a huge amount of energy to start. It was like trying to fill a bucket with a hole in the bottom; you have to pour water in very quickly just to keep it from running empty.

Additionally, the sides of these pillars were often rough, like a jagged rock. This caused the light to scatter and escape, making the "whisper" fade quickly.

The Solution: A Smooth Slide and a Better Roof

The team built a new version of these pillars with two essential improvements:

The Smooth Slide: They used a special chemical process to make the sides of the pillars perfectly smooth. Imagine a marble rolling down a polished glass slide instead of an uneven gravel path. This allowed the light to race along the edge without losing energy.
The Better Roof: They swapped the old mirrors for a new type made of different materials (Aluminum-Gallium-Arsenide). These new mirrors act like a "transparent window" for light entering, but like a "perfect mirror" for light trying to escape. This allowed them to send a laser beam straight through the center of the pillar to start the light, and then catch the laser beam shooting straight up from the top.

The Results: A Quiet, Efficient Laser

Thanks to these improvements, the new pillars worked incredibly well:

Low Power: They required very little energy to start laser operation. The paper mentions a threshold of just 240 microwatts (at a cold temperature of 130 Kelvin). To put this in perspective: previous methods required about 100 milliwatts. This is comparable to the difference between the energy of a tiny LED flashlight and a bright floodlight. They made the laser 400 times more efficient.
Multiple Colors: With pillars of different sizes, they saw that the light emerged in a "comb" pattern—several distinct colors (wavelengths) appeared simultaneously, like the teeth of a comb.
Single Color at Higher Temperatures: When they slightly warmed the pillar (to 130 Kelvin), the 5-micrometer-wide pillar settled down and began emitting only a single, pure color of laser light.
Stability: Even when they turned up the power, the color of the laser barely changed. It remained stable, which is crucial for deploying these in complex systems.

Why Is This Important? (According to the Paper)

The paper suggests that these tiny, efficient, surface-emitting lasers could be used to build arrays (grids) of lasers. Since they are so stable and can be tuned to specific colors by changing the size of the pillar, they could be used for a type of computing called Optical Reservoir Computing.

Imagine a choir. If you have a choir where every singer is slightly out of tune or requires a lot of energy to sing, the music is chaotic. But if you have a choir where every singer is perfectly tuned, consumes very little energy, and sings exactly the note you want, you can create complex, beautiful harmonies. The scientists believe these new pillars could act as perfect "singers" for future optical computers.

Summary

In short, the scientists built a better "light cage." By smoothing the walls and fixing the roof, they created a tiny laser that starts with very little energy, shoots straight up, and remains stable even when operated at higher power. This makes them much better candidates for future high-tech computing applications than the older, "leaky" versions.

1. Problem Statement

Semiconductor micropillars with vertical microcavities (formed by Distributed Bragg Reflectors, or DBRs) typically lase via axial modes. Although Whispering-Gallery Modes (WGMs)—which propagate along the active region and are confined by sidewalls—are well-known in microdisks and rings, their realization in micropillars presents specific challenges:

Excitation Efficiency: Previous WGM laser emission in micropillars often required excitation through the sidewalls, leading to very low light absorption by the gain medium (only a few percent).
Surface Emission Limitations: Although surface-emitting WGM laser emission (excitation and collection along the pillar axis) is desirable for applications such as optical reservoir computing (RC), it was historically limited to very low temperatures (4–14 K) and suffered from high threshold excitation powers (e.g., ~100 mW in earlier studies).
Thermal Instability: High pump powers frequently lead to thermal heating, causing mode shifts and line broadening, which destabilizes the laser stability required for dense arrays.

2. Methodology

The authors designed and fabricated high-quality micropillar cavities to overcome these limitations through specific structural and experimental optimizations:

Material System: The structure consists of InGaAs quantum dots (QDs) as the active region, embedded in a GaAs cavity.
Optimized DBRs: Instead of standard GaAs/AlGaAs pairs, the authors used low-absorbing Al $_{0.2}$ Ga $_{0.8}$ As/Al $_{0.9}$ Ga $_{0.1}$ As-DBRs. This choice minimizes optical absorption in the top mirror and enables efficient excitation and collection of emission along the pillar axis (vertical direction) without significant heating.
Fabrication: Micropillars with diameters of 2 to 7 $\mu$ m were produced using photolithography and dry etching. The process was optimized to produce smooth sidewalls with a vertical mesa profile (deviation < 1 degree), which is crucial for maintaining high Q-factors for WGMs.
Experimental Setup:
- Excitation: A continuous 808-nm laser diode was focused onto the pillar axis (top).
- Collection: Emission was collected along the same axis.
- Conditions: Measurements were performed in a cryostat at temperatures between 77 K and 130 K.
- Spot Size: A 7- $\mu$ m excitation spot was used to ensure uniform excitation density across various pillar diameters and to suppress vertical axial modes.

3. Main Contributions

First Low-Threshold Surface-Emitting WGM Laser Emission: The paper demonstrates WGM laser emission in micropillars for the first time, excited and collected via the pillar axial direction, with a significantly reduced threshold power.
Temperature Stability: Realization of single-mode WGM laser emission at 130 K, a significant improvement over the cryogenic temperatures (4–14 K) required in previous surface-emitting WGM studies.
Material Innovation: Successful implementation of low-absorbing Al-based DBRs to enable efficient vertical excitation while simultaneously suppressing thermal effects.
Application Readiness: Demonstration of spectrally uniform, low-threshold lasers suitable for optical reservoir computing (RC), where dense, diffractively coupled arrays are required.

4. Main Results

Laser Properties:
- Spectral Range: Laser emission was observed in the 930–970 nm range.
- Mode Structure:
  - Smaller pillars (2–4 $\mu$ m) showed single- or two-mode laser emission.
  - Larger pillars (6–7 $\mu$ m) exhibited a comb-like structure (simultaneous multimode laser emission) due to inhomogeneous broadening of the QD active region.
- Free Spectral Range (FSR): The separation between WGMs followed the expected inverse dependence on pillar diameter ( $FSR \propto 1/D$ ).
Threshold and Q-Factor:
- 77 K: The estimated threshold excitation power was ~180 $\mu$ W for a 2- $\mu$ m pillar and ~240 $\mu$ W for a 5- $\mu$ m pillar. The cold cavity Q-factor was estimated at ~10,000 for the 5- $\mu$ m pillar (rising to 13,400 with gain).
- 130 K: Single-mode laser emission was maintained for the 5- $\mu$ m pillar with a threshold of ~240 $\mu$ W and a cold cavity Q-factor of ~8,000.
Thermal Stability:
- In contrast to previous studies, the linewidth did not broaden significantly at high pump powers.
- Mode Shift: At 130 K, the emission mode showed a very stable redshift of only ~170 $\mu$ eV, even when the excitation power was increased 12-fold above threshold. This indicates that thermal effects are well controlled, likely due to the low-absorbing DBRs.
Spontaneous Emission Factor ( $\beta$ ): The $\beta$ -factor was calculated to be approximately 1.1–1.2 %, indicating efficient coupling of spontaneous emission into the lasing mode.

5. Significance

This work represents a crucial step towards the practical application of semiconductor microlasers in optical reservoir computing.

Scalability: The ability to excite and collect light vertically enables the fabrication of dense, spectrally homogeneous arrays of microlasers, which is a prerequisite for optical RC.
Efficiency: Reducing the threshold power from 100 mW (sidewall excitation) to **240 $\mu$ W** (surface emission) makes these devices practical for low-power integrated photonic circuits.
Robustness: The stability of emission wavelength and linewidth at elevated temperatures (up to 130 K) and high pump powers suggests that these devices can operate in less extreme cooling environments than previously thought possible for WGM microlasers.

In summary, the authors successfully developed a micropillar platform that combines high-quality WGMs with efficient surface emission, achieving low-threshold, thermally stable laser emission suitable for next-generation optical computer architectures.