🔬 materials science

Competition between vacancy creation and filling in defect-engineering of hBN

By irradiating freestanding monolayer hBN with ultra-low-energy Ar+ ions, this study reveals that boron single vacancies are the predominant defect formed contrary to simulation predictions, while impurity-driven vacancy filling plays a more significant role than previously assumed, complicating the selective engineering of quantum emitters.

Original authors: Shrirang Chokappa, Manuel Laängle, Barbara Maria Mayer, Vladimir Zoba\vc, Jacob Madsen, Diana Propst, David Lamprecht, Philipp Irschik, Arixin Bo, Clara Kofler, Vinzent Hana, Fabian Kraft, Clemens Man

Published 2026-09-11

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Shrirang Chokappa, Manuel Laängle, Barbara Maria Mayer, Vladimir Zoba\vc, Jacob Madsen, Diana Propst, David Lamprecht, Philipp Irschik, Arixin Bo, Clara Kofler, Vinzent Hana, Fabian Kraft, Clemens Mangler, Lado Filipovic, Toma Susi, Jani Kotakoski

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

Hexagonal boron nitride is a material that looks and feels a bit like a sheet of paper made of atoms, but it is incredibly strong and does not conduct electricity. For years, scientists have been fascinated by the idea that tiny flaws in this atomic sheet could act as quantum emitters, which are sources of light that could power future technologies like ultra-secure communication or advanced computers. However, a major hurdle has stood in the way: while researchers know these light sources come from missing atoms, or vacancies, in the crystal structure, they have struggled to control exactly which atoms go missing. Without the ability to create specific types of missing atoms on demand, it is nearly impossible to link a particular flaw to a specific color of light it emits. To solve this, scientists have tried shooting different kinds of particles at the material, but the results have often been unpredictable or difficult to see clearly.

A team of researchers recently took a fresh approach to this problem by firing a stream of argon ions at a single layer of hexagonal boron nitride. They used ions with very low energy, just enough to knock atoms out of place without shattering the entire sheet. Before they began, they prepared the material by heating it in a vacuum to ensure the surface was as clean as possible, removing any dust or chemical residue that might interfere with the experiment. They then exposed the clean sheets to the ion beam for a precise amount of time, aiming to create a controlled number of defects. After the bombardment, they examined the material using a powerful electron microscope that can see individual atoms, allowing them to count exactly what had changed.

The results surprised the team and challenged what computer simulations had predicted for years. Theoretical models had suggested that the low-energy ions would knock out nitrogen atoms more often than boron atoms, or perhaps create pairs of missing atoms. Instead, the researchers found that the ions overwhelmingly removed single boron atoms, leaving behind empty spots where boron used to be. These single boron vacancies were the most common defect, appearing roughly twice as often as double vacancies where two atoms were missing. This discovery suggests that the way these defects form is more selective than previously thought, favoring the removal of boron over nitrogen under these specific conditions.

However, the story did not end with simple empty spots. As the researchers looked closer, they discovered a second, competing process that had been largely overlooked. While the ions were knocking atoms out, other atoms floating around in the environment were rushing in to fill the gaps. Specifically, silicon atoms from the surrounding contamination were jumping into the empty boron spots. This vacancy filling happened so frequently that it created a significant number of new defects that looked different from the original empty spots. The researchers found that for every few vacancies created, a silicon atom would often take its place, effectively hiding the original damage. This means that simply counting the holes left by the ions is not enough; one must also account for the new atoms that sneak in to patch the holes.

The team also compared their findings to what happens when similar ions hit graphene, a material made entirely of carbon. In graphene, the ions tend to create pairs of missing atoms, and impurities do not fill the gaps as readily. In hexagonal boron nitride, the behavior is different: single missing atoms are more common, and the filling of those gaps by impurities is a major factor. This difference highlights that the rules governing how defects form in this material are unique and more complex than in other two-dimensional materials. The researchers noted that their measurements showed a clear preference for boron vacancies, a finding that contradicts earlier simulations which predicted a mix of nitrogen and boron vacancies or a dominance of nitrogen vacancies.

Despite the complexity introduced by the filling atoms, the study confirms that using very low-energy ions is a highly efficient way to create defects in hexagonal boron nitride. The process created a density of defects that was high enough to be useful for engineering purposes but low enough that the ions mostly hit pristine areas of the material rather than piling up on the same spot. The researchers concluded that while the ability to create specific defects is promising, the uncontrollable filling of those defects by impurity atoms like silicon presents a significant challenge. If scientists want to build reliable quantum devices using this material, they will need to find a way to prevent these stray atoms from filling the vacancies, or at least learn to control that process as precisely as they control the creation of the holes themselves. This work provides a clearer map of the terrain, showing that the path to controlling quantum emitters in this material involves not just knocking atoms out, but also managing the chaotic environment that tries to fill them back in.

Technical Summary: Competition between Vacancy Creation and Filling in Defect-Engineering of hBN

Problem Statement
Hexagonal boron nitride (hBN) is a promising solid-state host for quantum emitters, yet a definitive correlation between specific point defect structures and their emission spectra remains elusive. While various irradiation methods (lasers, ions, neutrons, electrons) can generate these emitters, controlling the specific type of defect created is challenging. Previous analytical potential molecular dynamics simulations suggested that ultra-low-energy noble gas ion irradiation (specifically <200 eV Ar+) should offer high selectivity, potentially creating nitrogen single vacancies ( $V_N$ ) or a mix of boron ( $V_B$ ) and nitrogen vacancies with high efficiency. However, these predictions lacked experimental confirmation. Furthermore, the role of impurity atoms (such as carbon and silicon) in filling vacancies during or after irradiation was not well understood, potentially obscuring the true defect distribution.

Methodology
The authors prepared freestanding, monolayer hBN samples grown via chemical vapor deposition (CVD) on copper foil. Using an electrochemical delamination method, the hBN was transferred to Quantifoil gold TEM grids with an amorphous carbon support. To ensure cleanliness, samples underwent a rigorous cleaning process involving solvent baths (acetone and isopropyl alcohol) followed by baking in ultra-high vacuum (UHV) at 250°C for 180 minutes to remove surface contamination.

The samples were irradiated with singly charged argon ions ( $Ar^+$ ) at an ultra-low energy of approximately 150 eV, generated by a microwave plasma source. The ion energy profile was characterized using a Faraday cup, confirming a mean energy between 149–155 eV. The average ion fluence was controlled at $0.8 \pm 0.2$ ions nm $^{-2}$ .

Defect characterization was performed using Scanning Transmission Electron Microscopy (STEM) with a medium-angle annular dark-field (MAADF) detector at 60 kV in UHV. To minimize electron-beam-induced damage, a semi-automated imaging tool was employed to acquire atomic-resolution images of predefined regions without overlapping frames. A Convolutional Neural Network (CNN)-assisted workflow was used to identify and classify defects based on atomic number (Z)-contrast.

Complementary atomistic simulations were conducted using Ehrenfest molecular dynamics (via the GPAW software) to model 150 eV $Ar^+$ collisions with hBN. Additionally, Density Functional Theory (DFT) simulations were performed to investigate the energetics of carbon and oxygen adatoms filling boron and nitrogen vacancies.

Key Results

Defect Density and Distribution: The intrinsic defect density of the cleaned, pre-irradiation samples was measured at $0.021$ nm $^{-2}$ . After irradiation, the total defect density increased to $0.228$ nm $^{-2}$ , corresponding to an ion-induced defect density of approximately $0.207$ nm $^{-2}$ .
Dominance of Boron Vacancies: Contrary to simulation predictions that suggested nitrogen single vacancies ( $V_N$ ) would be prominent or that boron and nitrogen vacancies would form with equal probability, the experiments revealed a strong predominance of boron single vacancies ( $V_B$ ). The post-irradiation defect distribution showed $V_B$ at a density of $0.089$ nm $^{-2}$ (39% of total defects), followed by double vacancies ( $V_{BN}$ ) at $0.042$ nm $^{-2}$ (18.6%). Nitrogen single vacancies ( $V_N$ ) were significantly less frequent ($0.018$ nm $^{-2}$ , 7.9%).
Vacancy Filling by Impurities: A significant and unexpected finding was the high density of silicon atom substitutions in boron sites ( $Si_B$ ), measured at $0.035$ nm $^{-2}$ (15.5% of total defects). The authors observed that these impurity atoms preferentially occupied boron sites rather than nitrogen sites. DFT simulations supported this, showing a higher likelihood for carbon adatoms to fill boron vacancies compared to nitrogen vacancies. The data suggests that impurity atoms (likely mobilized from surface contamination during irradiation) fill vacancies immediately after or during their creation, potentially masking the true number of vacancies created.
Discrepancy with Simulations: The experimental results contradict previous analytical potential molecular dynamics simulations, which predicted a higher probability for $V_N$ or equal $V_B$ / $V_N$ creation. The authors note that standard simulations may fail to capture the complex dynamics of impurity filling and charge localization effects in insulating hBN.

Significance and Claims
The paper claims to demonstrate that ultra-low-energy noble gas ion irradiation is an effective method for defect engineering in hBN, achieving a high degree of selectivity for boron single vacancies. However, this selectivity is complicated by a competing mechanism: the filling of vacancies by impurity atoms (specifically silicon and carbon) which are mobilized during the irradiation process.

The authors emphasize that their results challenge existing theoretical models that do not account for impurity filling or the specific charge dynamics in insulating hBN. They conclude that while defect production is significantly more efficient with <1 keV ions compared to higher-energy ions, the interpretation of defect creation must carefully consider the role of contamination and vacancy filling. The study highlights that the "clean" creation of specific defects is difficult to achieve without controlling the impurity environment, and that previous assumptions about defect distributions in hBN may require revision based on these experimental findings.

More like this