🔬 mesoscale physics

Localized Excitonic Emission in Wafer-Scale MOCVD-Grown GaSe 2D Nanosheets for Classical and Non-Classical Light Sources

This study demonstrates the wafer-scale MOCVD growth of 2D GaSe nanosheets, revealing that defect-induced localized emission enables both broad classical light and single-photon quantum emission, thereby establishing a scalable platform for integrated photonic technologies.

Original authors: Bhabani Sankar Sahoo, Nils Fritjof Langlotz, Shachi Machchhar, Kartik Gaur, Robin Günkel, Max Bergmann, Naghmeh Ghadghooni, Aris Koulas-Simos, Jürgen Belz, Chirag Chandrakant Palekar, Maximilian Ries

Published 2026-05-25

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Bhabani Sankar Sahoo, Nils Fritjof Langlotz, Shachi Machchhar, Kartik Gaur, Robin Günkel, Max Bergmann, Naghmeh Ghadghooni, Aris Koulas-Simos, Jürgen Belz, Chirag Chandrakant Palekar, Maximilian Ries, Kerstin Volz, Stephan Reitzenstein, Imad Limame

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

Imagine you are trying to build a city of tiny, ultra-thin light-emitting buildings (2D materials) on a massive, flat foundation (a silicon wafer). For years, scientists have been great at building these cities with one specific type of material (like the "Transition Metal Dichalcogenides"), but they've mostly been using a method that's like hand-carving each brick one by one. This is slow, messy, and you can't build a whole city with it.

This paper is about a new way to build a different type of material called Gallium Selenide (GaSe) using a method called MOCVD. Think of MOCVD as a high-tech "spray paint" or "fog machine" that can coat an entire city-sized wafer with this material at once, layer by layer, in a very controlled way.

Here is the story of what the researchers found, broken down simply:

1. The "Spray Paint" Experiment

The team used this "fog machine" to grow GaSe on a special silicon-based foundation. They ran the machine for two different amounts of time to see what would happen:

The Short Spray (3 minutes): This created very thin, patchy islands of material, like a few scattered puddles of paint.
The Long Spray (30 minutes): This created a thick, continuous blanket of material, covering the whole surface like a thick layer of snow.

2. What the "Thin" vs. "Thick" Layers Looked Like

When they looked closely at these layers under powerful microscopes:

The Thick Layer (30 mins): It was a bit messy. It had lots of bumps and imperfections. When they shined light on it, it glowed with a broad, fuzzy rainbow of colors. It was like a lightbulb that was slightly out of focus; the light was there, but it wasn't sharp or specific.
The Thin Layer (3 mins): This was much more interesting. Because the layer was so thin and patchy, the light got "trapped" in tiny, specific spots. Instead of a fuzzy rainbow, these spots glowed with sharp, distinct colors (like a laser pointer).

3. The "Quantum" Surprise

The most exciting part happened with the thin, 3-minute sample. The researchers found that some of those tiny, sharp glowing spots were behaving in a very strange, "quantum" way.

Usually, when a light source glows, it shoots out many photons (particles of light) at once, like a hose spraying water. But these specific spots were acting like a single-fire gun. They were shooting out one single photon at a time, waiting for the first one to leave before sending the next.

They proved this by measuring the light and finding a value (called $g^{(2)}(0)$ ) of 0.15. In the world of quantum physics, anything below 0.5 is a clear sign that you have a "single-photon source." This is the kind of light needed for future ultra-secure communication and quantum computers.

4. Why Did This Happen? (The "Defect" Secret)

You might think that "defects" (imperfections) in a material are bad. Usually, they are. But in this case, the researchers found that the imperfections were actually the heroes.

Think of the material like a bumpy trampoline.

In the thick sample, the trampoline was so bumpy and chaotic that the light (the ball) bounced everywhere, creating a messy, broad glow.
In the thin sample, the "bumps" (defects) created tiny, deep valleys. The light got stuck in these valleys. Because the light was trapped in such a small, isolated spot, it could only escape one particle at a time.

The paper concludes that these "defect-induced" traps are actually a feature, not a bug. They naturally created the perfect conditions for single-photon emission without needing to build complex, expensive structures to force it to happen.

The Bottom Line

The researchers successfully grew a whole wafer of this material using a scalable, industrial method (MOCVD). They discovered that by controlling how long they grew it, they could create:

Thick layers that act like standard, bright light sources (good for classical tech).
Thin layers that naturally form tiny "traps" which emit single photons (good for quantum tech).

This is a big deal because it shows you can make these high-tech quantum light sources on a large scale, using a method that fits with existing silicon technology, rather than having to hand-craft them one by one. The "imperfections" in the thin layers turned out to be the secret sauce for creating quantum light.

Technical Summary: Localized Excitonic Emission in Wafer-Scale MOCVD-Grown GaSe 2D Nanosheets

Problem Statement
The integration of two-dimensional (2D) semiconductors into photonic technologies is hindered by the lack of scalable, controllable growth methods. While transition metal dichalcogenides (TMDCs) have been extensively studied for their tunable band structures and excitonic properties, III–VI monochalcogenides like gallium selenide (GaSe) remain underexplored in this context. Existing GaSe research relies heavily on mechanical exfoliation, which limits lateral size and reproducibility, or conventional chemical vapor deposition (CVD), which often struggles with uniformity and phase purity. There is a critical need for wafer-scale growth techniques that can produce high-quality GaSe layers compatible with silicon-based platforms for both classical and quantum light sources.

Methodology
The authors employed Metal–Organic Chemical Vapor Deposition (MOCVD) to grow GaSe layers on GaP/Si substrates, a platform chosen for its lattice compatibility and established role in silicon photonics.

Growth Process: Using triethylgallium (TEGa) and diisopropylselenide (DiPSe) precursors, two samples were synthesized with varying growth times (3 minutes and 30 minutes) to control thickness, ranging from a few monolayers to several micrometers. A thin GaP buffer layer was deposited prior to GaSe growth to stabilize the surface.
Structural Characterization: The samples were analyzed using Atomic Force Microscopy (AFM), Scanning Electron Microscopy (SEM), X-ray Diffraction (XRD), Raman spectroscopy (including spatial mapping), and High-Resolution Scanning Transmission Electron Microscopy (STEM) to assess morphology, crystallinity, phase purity, and epitaxial alignment.
Optical Characterization: The optical properties were investigated via micro-photoluminescence (µPL), time-resolved photoluminescence (TRPL), and cathodoluminescence (CL) mapping at low temperatures (20 K).
Quantum Optical Analysis: To verify the nature of the emission, second-order autocorrelation measurements ( $g^{(2)}(\tau)$ ) were performed using a Hanbury Brown and Twiss (HBT) interferometer on the 3-minute sample.

Key Results

Structural and Morphological Evolution:
- 3-Minute Sample: Exhibited anisotropic, vertically oriented nanosheets growing along the $\langle 111 \rangle$ directions of the GaP substrate, with heights up to 262 nm. STEM confirmed an atomically abrupt interface and a $\gamma'$ stacking sequence.
- 30-Minute Sample: Showed a transition to planar, trigonal flakes on a predominantly flat GaP (100) surface, indicating a shift toward laterally extended growth.
- Phase Purity: Raman and XRD analyses confirmed predominantly phase-pure GaSe. While weak Raman features suggested potential $Ga_2Se_3$ in the thicker sample, XRD showed no secondary phase reflections, suggesting these features arose from local laser-induced effects rather than bulk phase impurities.
Optical Emission Characteristics:
- 30-Minute Sample: Displayed intense, broad photoluminescence (PL) spanning 1.7–2.0 eV. Cathodoluminescence (CL) mapping revealed spatially localized emission spots superimposed on a broad background, attributed to increased structural disorder and defect-mediated recombination in thicker layers.
- 3-Minute Sample: Exhibited discrete, narrow emission lines with significant spatial localization. CL mapping identified isolated bright spots (~300 nm) with an emitter density of $8 \times 10^7$ cm $^{-2}$ . Spectral analysis revealed narrow peaks (FWHM ~20 meV) centered at 1.91 eV.
Defect-Induced Localization:
- Power-dependent µPL showed a systematic redshift and linewidth broadening with increasing excitation power, consistent with carrier-induced screening of localized potentials and bandgap renormalization.
- Temperature-dependent analysis revealed a power-law behavior ( $I_{PL} \propto P^\alpha$ ) where the exponent $\alpha$ approached unity at higher temperatures, indicating thermal activation of excitons out of localized states.
- Polarization-resolved measurements showed anisotropic optical transitions with a degree of linear polarization (DLP) of ~0.5.
- TRPL measurements yielded an excited-state lifetime of $1.26 \pm 0.01$ ns, intermediate between MBE-grown and exfoliated GaSe, suggesting a distinct confinement regime governed by defects.
Single-Photon Emission:
- HBT measurements on a localized emitter in the 3-minute sample yielded a second-order correlation function of $g^{(2)}(0) = 0.15 \pm 0.10$ at 4 K. This value is well below the 0.5 threshold, confirming the emission of single photons. The authors note this purity is achieved without encapsulation, though it is slightly higher than values reported for encapsulated MBE-grown systems, likely due to surface charge noise.

Significance and Claims
The paper establishes wafer-scale MOCVD-grown 2D GaSe as a viable platform for both classical and non-classical light sources. The primary contribution is demonstrating that growth duration can be used to engineer the structural morphology and, consequently, the optical emission properties of GaSe.

The authors claim that the observed narrow, spatially localized emission and single-photon generation in the thin (3-minute) sample are driven by defect-induced localization rather than intrinsic excitonic recombination. This finding highlights "defect-engineered localization" as a route toward scalable quantum photonics. Unlike previous works relying on mechanical exfoliation or complex strain engineering, this study shows that high single-photon purity can be achieved through controlled MOCVD growth on silicon-compatible substrates. The work positions MOCVD-grown GaSe as an intermediate step between exfoliated systems and fully engineered heterostructures, offering a pathway for scalable integration of layered semiconductors into silicon-based photonic architectures. Future improvements in spectral stability and photon purity are suggested to be achievable through surface passivation or dielectric encapsulation.

1. The "Spray Paint" Experiment

2. What the "Thin" vs. "Thick" Layers Looked Like

3. The "Quantum" Surprise

4. Why Did This Happen? (The "Defect" Secret)

The Bottom Line

More like this