🔬 optics

Deterministic generation of single B centers in hBN by one-to-one conversion from UV centers

This study utilizes in-situ cathodoluminescence monitoring to demonstrate the deterministic, one-to-one conversion of ubiquitous UV centers into single B centers in hexagonal boron nitride, enabling the controlled creation of single-emitter arrays and providing new insights into their activation mechanism.

Original authors: Andrés Núñez Marcos, Christophe Arnold, Julien Barjon, Stéphanie Buil, Jean-Pierre Hermier, Aymeric Delteil

Published 2026-03-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Andrés Núñez Marcos, Christophe Arnold, Julien Barjon, Stéphanie Buil, Jean-Pierre Hermier, Aymeric Delteil

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 have a block of hexagonal boron nitride (hBN). Think of this material not as a boring rock, but as a vast, empty city made of tiny atomic bricks. In this city, there are "ghosts" hiding in the walls. These ghosts are called color centers, and they are special because they can glow and shoot out single particles of light (photons), which is the holy grail for building future quantum computers.

There are two main types of ghosts in this city:

The UV Ghosts: They are everywhere, but they glow in the ultraviolet (invisible to our eyes). They are like the "default" state of the city.
The Blue Ghosts (B Centers): These are the VIPs. They glow in the blue spectrum and have amazing superpowers for quantum technology. But they don't just appear; they have to be "awakened."

The Problem: Shooting in the Dark

For a long time, scientists wanted to create these Blue Ghosts on demand. They knew that if they shot a beam of electrons (like a tiny, invisible laser pointer) at the material, the UV Ghosts would sometimes turn into Blue Ghosts.

However, it was a lottery.

You'd shoot the electron beam at a spot.
You'd hope a Blue Ghost appeared.
But you had no way of knowing in the moment if you got one, or if you accidentally got two, or none at all.
It was like trying to bake a cake by throwing ingredients into a dark oven and hoping you didn't burn the house down or forget the eggs.

The Breakthrough: The "Magic Mirror"

The team at Université Paris-Saclay built a special setup that acts like a magic mirror with two eyes.

Eye 1 (UV Filter): Watches for the UV Ghosts.
Eye 2 (Blue Filter): Watches for the Blue Ghosts.

They put this setup inside a Scanning Electron Microscope (SEM). As they scanned the material with the electron beam, they watched the ghosts in real-time.

The Discovery: A Perfect Swap

What they saw was like watching a dance partner switch.

When the electron beam hit a spot, a UV Ghost would suddenly vanish.
At the exact same moment, a Blue Ghost would pop into existence right where the UV one was.
It wasn't random. It was a 1-to-1 swap. One UV Ghost turned into exactly one Blue Ghost.

They even saw the reverse happen! Sometimes, a Blue Ghost would turn back into a UV Ghost. This proved that the two ghosts are actually the same "soul" wearing different outfits, just rearranged by the electron beam.

The Analogy: Imagine a room full of people wearing red hats (UV Ghosts). You walk in with a magic wand (the electron beam). When you tap someone, their red hat instantly transforms into a blue hat (Blue Ghost). You don't just get a blue hat; the red hat disappears completely. It's a perfect trade.

The Solution: The "Stop" Button

Because they could see this swap happen instantly, they invented a new way to build quantum devices: The "Heralded" Method.

Here is how they built a perfect grid of single Blue Ghosts:

Aim: They pointed the electron beam at a specific spot.
Watch: They waited for the "pop" signal (the moment the UV light died and the Blue light appeared).
Stop: The instant they saw the Blue Ghost appear, they turned off the electron beam.

This is like a chef tasting a soup. Instead of cooking for a fixed time and hoping it's done, they taste it every second. The moment it's perfect, they take it off the heat.

Fixing Mistakes: The "Eraser"

Sometimes, the beam might accidentally turn two UV Ghosts into Blue ones at the same spot. No problem!

They used a high-power laser to "photobleach" (erase) the extra Blue Ghost.
They watched the light dim until only one Blue Ghost remained.
Now, they had a guaranteed single emitter.

Why Does This Matter?

Solving the Mystery: They figured out what these ghosts actually are. The UV Ghost is a pair of carbon atoms lying flat (horizontal). The Blue Ghost is the same pair, but standing up (vertical), with a tiny hole (vacancy) left behind. The electron beam flips them over and kicks out a loose atom to lock them in place.
Perfect Manufacturing: Before this, making quantum devices was like rolling dice. Now, it's like using a 3D printer. They can place exactly one quantum light source wherever they want, with 100% success.

In a nutshell: This paper is about learning to watch a magic trick in real-time, figuring out exactly how the trick works, and then using that knowledge to perform the trick perfectly every single time, creating the building blocks for the quantum internet.

1. Problem Statement

Hexagonal boron nitride (hBN) hosts various quantum emitters, with "B centers" (blue-emitting, ~436 nm) being particularly promising for quantum photonics due to their excellent optical properties and ability to be locally activated by an electron beam. However, two major challenges have hindered their deterministic integration into photonic devices:

Stochastic Activation: Without real-time monitoring, the electron-beam activation process yields an unpredictable number of emitters per site (following a Poisson distribution), making it difficult to create arrays with exactly one emitter per location.
Unclear Mechanism: The microscopic structure of B centers and the precise mechanism of their activation from precursor defects remain debated. Previous studies relied on ensemble measurements at room temperature, obscuring the dynamics of individual defects and their relationship to ubiquitous UV-emitting centers (4.1 eV).

2. Methodology

The authors developed a specialized experimental setup to monitor individual color centers in real-time during electron irradiation:

Experimental Setup: A scanning electron microscope (SEM) equipped with a parabolic mirror for cathodoluminescence (CL) collection. The emitted light is split into two spectral channels:
- UV Channel: Detects UV centers (4.1 eV emission).
- Blue Channel: Detects B centers (~436 nm emission).
- Both channels utilize avalanche photodiodes (APDs) for single-photon sensitivity.
In-situ Monitoring: The team performed sequential CL scans (1.6 µm × 1.9 µm) on hBN flakes containing native carbon impurities. They monitored the temporal evolution of individual emitters to observe activation/deactivation events as the electron beam scanned the sample.
Deterministic Fabrication Protocol:
1. Heralded Activation: Irradiate a specific spot and monitor the blue channel. Upon detecting the first discrete "upward step" in photon counts (indicating a single B center creation), the electron beam is immediately blanked.
2. Heralded Deactivation (Correction): If a site accidentally generates two emitters (double activation), a high-power continuous-wave (CW) laser is used to selectively photobleach one emitter while monitoring the photoluminescence (PL) signal. The process stops once a "downward step" in intensity is observed, leaving a single emitter.
Characterization: The resulting arrays were characterized using confocal photoluminescence (PL) and Hanbury Brown and Twiss (HBT) interferometry to measure the second-order correlation function, $g^{(2)}(0)$ , to verify single-photon emission.

3. Key Results

One-to-One Conversion: The study provides direct evidence of a spatially and temporally correlated one-to-one conversion between individual UV centers and B centers. When a UV center deactivates, a B center activates at the exact same location almost instantaneously (faster than the 100 ms pixel integration time). Conversely, the process is partially reversible, with B centers converting back to UV centers under specific conditions.
Microscopic Insight: The observed interconversion supports the hypothesis that B centers are vertical carbon dimers in a split-interstitial configuration, while UV centers are in-plane carbon dimers ( $C_B C_N$ ). The authors propose a mechanism where the electron beam induces a structural reorientation of the dimer, creating a vacancy ( $V_B$ or $V_N$ ) in the process:
$C_B C_N \rightleftharpoons (2C_i)_N + V_B$
The subsequent migration of the mobile vacancy explains the long-term stability of the B center state.
Deterministic Array Generation:
- Using the "heralded activation" technique, the authors created a 4×2 array of emitters.
- Success Rate: 7 out of 8 sites contained exactly one emitter ( $g^{(2)}(0) < 0.5$ ).
- Correction: The 8th site initially showed double emission ( $g^{(2)}(0) \approx 0.67$ ). By applying the "heralded photobleaching" technique, they successfully reduced this to a single emitter ( $g^{(2)}(0) \approx 0.29$ ).
- This approach achieves a 100% yield of single emitters per site, a significant improvement over the theoretical ~37% limit of stochastic Poissonian activation.

4. Key Contributions

Technological: Development of a dual-channel (UV/Blue) in-situ CL monitoring system capable of detecting single-emitter dynamics, enabling deterministic "top-down" fabrication of quantum emitter arrays.
Methodological: Introduction of a two-step protocol (heralded activation + heralded photobleaching) to correct stochastic errors, ensuring high-fidelity single-emitter arrays.
Scientific: Resolution of the microscopic relationship between UV and B centers. The observation of direct interconversion strongly validates the structural models of UV centers as in-plane carbon dimers and B centers as vertical carbon dimers, resolving long-standing debates in the field.

5. Significance

This work represents a major step forward in the practical application of hBN for quantum technologies. By overcoming the stochastic nature of defect activation, the authors provide a reliable pathway for the deterministic integration of single-photon sources into nanophotonic devices. This is crucial for scaling up quantum networks and quantum computing architectures that require precise placement of identical quantum emitters. Furthermore, the elucidation of the activation mechanism offers a deeper understanding of defect engineering in 2D materials, potentially guiding the optimization of other quantum defects in wide-bandgap semiconductors.