Quantum sensing using optically addressable spin defects in 3D crystals (like diamond NV centers) and 2D van der Waals (vdW) materials (like hBN) is rapidly advancing nanoscale measurements of magnetic fields, temperature, and other stimuli. However, diamond fabrication is challenging and costly, and NV center properties degrade near surfaces. 2D hBN offers advantages due to its stable spin defects in atomically thin flakes. Most hBN studies focus on ensembles of negatively charged boron vacancies (V−), which are limited by the optical diffraction limit and their intrinsic spin axis (S=1) making them insensitive to off-axis magnetic fields. Zero-dimensional (0D) materials, like quantum dots, have been used but suffer from short spin lifetimes at room temperature. One-dimensional (1D) vdW nanotubes with optically active spin defects offer unique opportunities due to their small size, absence of dangling bonds, and potential for coupling to mechanical vibrations. While spin qubits in carbon nanotubes have been observed through electronic transport and EPR at low temperatures, optically detected magnetic resonance (ODMR) – crucial for quantum sensing – remains elusive. This paper reports the observation of single optically active spin defects in BNNTs at room temperature, opening new avenues for omnidirectional and high-resolution quantum sensing.

The literature review highlights the limitations of current quantum sensing technologies. Diamond NV centers, while effective, are costly and challenging to fabricate, with performance degrading near surfaces. While 2D hBN offers improvements with stable spin defects in atomically thin flakes, studies primarily focus on ensembles of V− defects, limited by diffraction and anisotropic sensitivity due to their S=1 spin state. Quantum dots, despite showing promise, suffer from short spin lifetimes at room temperature. The study acknowledges previous work on spin qubits in carbon nanotubes but emphasizes the lack of ODMR observation, a critical gap for quantum sensing applications. This paper builds upon this existing research by exploring the potential of 1D BNNTs for overcoming these limitations.

The study involved several key steps. First, single optically active spin defects in BNNTs (average diameter 50 nm) were identified and characterized using a home-built confocal microscope. ODMR measurements revealed negligible zero-field splitting (ZFS), suggesting a spin S=1/2 ground state and lack of an intrinsic quantization axis, critical for omnidirectional sensing. Rabi oscillation measurements further confirmed the S=1/2 ground state. To enhance the probability of finding spin defects, BNNTs were treated with carbon ion implantation and thermal annealing. CW ODMR measurements were performed at various magnetic field orientations to demonstrate omnidirectional sensitivity, observing consistent resonance frequency and contrast regardless of rotation. Pulsed ODMR measurements determined spin relaxation (T1) times and coherence times. To showcase omnidirectional sensing, the anisotropic magnetization of a Fe3GeTe2 (FGT) flake was measured at varying temperatures. A method for deterministically transferring a single BNNT with spin defects onto an AFM cantilever was developed for scanning probe magnetometry. The stray magnetic field distribution near nickel patterns was mapped by scanning the BNNT-AFM probe, with results validated by simulations. Detailed descriptions of sample preparation, confocal microscopy, ODMR measurements (CW and pulsed), and scanning probe magnetometry are provided in the methods section.

The study's key findings include the observation of single optically active spin defects in BNNTs at room temperature. These defects exhibit a spin S=1/2 ground state with negligible ZFS, resulting in orientation-independent magnetic field sensing. This omnidirectional sensitivity was demonstrated by measuring the anisotropic magnetization of a Fe3GeTe2 flake at different temperatures, observing a significant difference between out-of-plane and in-plane magnetization below the Curie temperature. Scanning probe magnetometry, employing a BNNT attached to an AFM cantilever, achieved high-resolution mapping of stray magnetic fields near nickel patterns. The spatial resolution was limited by the probe-sample distance (a few hundred nanometers), but the use of contact mode and feedback stabilization is suggested for future nanometer-scale resolution. The sensitivity of the BNNT spin defects was estimated to be around 80 µT/Hz on a gold stripline, with the best sensitivity achieved being 21 µT/Hz. For the AFM cantilever setup, the sensitivity was 245 µT/Hz. The BNNTs proved more robust than diamond NV scanning probes, potentially enabling contact mode operation.

The findings address the limitations of current quantum sensing technologies by demonstrating a novel approach utilizing BNNTs with their inherent omnidirectional sensitivity and robustness. The S=1/2 spin state, lacking an intrinsic quantization axis, overcomes the directional limitations of existing S=1 defects like NV centers and V− centers in hBN. The development of a deterministic transfer method for attaching BNNTs to AFM cantilevers enables high-resolution scanning probe magnetometry, surpassing the limitations of optical diffraction. The robustness of BNNTs also allows for potential contact-mode scanning, opening possibilities for in-situ characterization of samples without damage. These results are significant for advancing quantum sensing in various fields, including condensed matter physics and biology, owing to BNNTs' small size and potential for cellular insertion.

This work successfully demonstrated the use of single optically addressable spin defects in BNNTs for omnidirectional and high-resolution magnetic field sensing. The unique properties of these S=1/2 defects, combined with the robust nature and small size of BNNTs, provide a promising platform for atomic-scale quantum sensing applications. Future work should focus on improving the sensitivity through optimizing defect selection and implementing feedback stabilization for nanometer-scale resolution in scanning probe magnetometry, and exploring applications in biological systems.

The current study has some limitations. The yield of spin defects in untreated BNNTs was relatively low (approximately 5%), requiring ion implantation and annealing to increase the yield to 20–30%. The spatial resolution of the scanning probe magnetometry was limited by the probe-sample distance (a few hundred nanometers). Further improvement requires implementing feedback stabilization and reducing the separation. The sensitivity of the BNNT spin defects varied from defect to defect. While improvements have been made through defect selection, further optimizations are needed.