Gold nanopillars on a microwave waveguide combine plasmonics and strain to push hBN quantum sensor sensitivity near the best values on record.
(Nanowerk Spotlight) Every electronic device, every living organism, and every mineral deposit produces a magnetic field. Most are too weak for conventional instruments to pick up. Quantum sensors can detect them by using atomic-scale defects in crystals whose electron spins respond to magnetic influences and signal that response as a change in emitted light.
The most established version relies on nitrogen-vacancy centers in diamond, but diamond is a bulky, rigid crystal. The defect that does the sensing remains physically distant from whatever is being measured, and that distance introduces noise and reduces sensitivity.
Hexagonal boron nitride (hBN) takes a different approach. A layered material that can be thinned to just a few atomic layers, hBN can be placed directly onto surfaces, wrapped around nanostructures, or built into two-dimensional quantum sensing chips. Its negatively charged boron vacancy defects, which can be created via helium ion bombardment, respond to magnetic fields, temperature, and strain, and can be read out at room temperature using optically detected magnetic resonance (ODMR). This technique pairs a laser with a microwave field to produce a measurable dip in fluorescence at the defect’s spin resonance frequency.
The limitation has been brightness. Boron vacancy defects emit light through an inherently inefficient process, producing a weak optical signal that caps sensor performance. Previous efforts coupled hBN with metallic films, nanotrenches, or dielectric nanopillars, each boosting photoluminescence to some extent. Yet none delivered a meaningful improvement in DC magnetic field sensitivity, which depends simultaneously on photoluminescence intensity, ODMR contrast, and resonance linewidth.
A study published in Advanced Materials (“Coupling Nanostructured Plasmon–Strain Microwave Waveguide to Spin Defects in Hexagonal Boron Nitride for High‐Sensitivity Quantum Sensors”) now presents a device architecture that improves all three factors simultaneously. It matches the magnetic field sensitivity previously achieved only with costly isotopically purified hBN crystals, but does so through waveguide design alone, using standard semiconductor fabrication, operating at room temperature, and requiring no elaborate tuning of laser or microwave parameters.
That combination makes the platform practical for integration into chip-scale devices. The research team fabricated arrays of gold nanopillars, each roughly 520 nm across and 125 nm tall, onto the narrow central strip of a single-port gold coplanar waveguide, a type of microwave transmission line designed to deliver radiofrequency energy efficiently to the defects. A 5 nm coating of alumina, deposited by atomic layer deposition, covered both the flat and pillared regions before an hBN flake containing boron vacancy defects was transferred on top.
(a) Schematic of the quantum-sensing prototype device featuring a nanostructured plasmon-strain coplanar (gold) waveguide coupled to an SMA connector for microwave driving. The inset shows a magnified view of the hBN, coupled with PNRs. (b) Atomic structure of a boron vacancy (VB−) embedded in the hexagonal lattice of hBN. (c) Illustration of the three waveguide configurations used in this study: flat gold (bare gold without Al2O3 coating), off-PNR (flat gold with Al2O3 coating), and on-PNR (gold pillars with Al2O3 coating). (d) FESEM image of an hBN flake deterministically transferred onto the nanostructured (PNR) waveguide. (e) Confocal PL map of VB− defects acquired from the area outlined by the red square in Figure 1d. The red and blue circles denote the off- and on-PNR sites, respectively. (f) Energy level structure of the VB− center, consisting of a triplet ground and excited states. (g) Schematic representation of the hBN lattice with high symmetry (unstrained) with D3h point and low symmetry (strained) configurations. (Image: Reproduced from DOI:10.1002/adma.202516761, CC BY) (click on image to enlarge)
Each component targets a different bottleneck. The gold nanopillars act as plasmonic nanoresonators, concentrating electromagnetic fields at both the 532 nm excitation wavelength and the roughly 800 nm emission wavelength. This dual resonance amplifies defect excitation and helps the defects radiate more efficiently. The alumina spacer prevents fluorescence quenching, the energy loss that occurs when light emitters sit directly on a metal surface.
Strain provides the third enhancement channel. When the thin hBN flake draped over the protruding pillars, it developed in-plane tensile strain of approximately 0.23%, confirmed independently by Raman spectroscopy and atomic force microscopy. This deformation breaks the symmetry of the boron vacancy’s local lattice environment, altering the spacing of its spin energy levels and changing the rates at which electrons cross between spin states. The net effect is faster light emission and a stronger spin-dependent contrast in the ODMR signal.
Photoluminescence at sites on the nanopillars exceeded 5 million counts per second, roughly tenfold higher than on bare gold. ODMR contrast reached approximately −17% on the pillars, compared with −11% on flat alumina-coated gold. Previous strain-only experiments using silicon dioxide nanopillars had achieved contrasts of just −0.5% to −1.4%, placing the new result more than an order of magnitude ahead. Measurements at four points across the flake showed that both photoluminescence and ODMR contrast rose steadily with local tensile strain, confirming a direct link between mechanical deformation and sensor performance.
At the best measurement point, the device reached a DC magnetic field sensitivity of 9.4 µT/√Hz, compared with 91.3 µT/√Hz on bare gold and 33.6 µT/√Hz on flat alumina-coated regions. That nearly tenfold improvement over bare gold was achieved without extensive optimization of laser power or microwave conditions.
One trade-off emerged from the curved geometry. Because the hBN bends over the nanopillars, the boron vacancy defects tilt at slightly different angles relative to the microwave field, producing nonuniform driving across the ensemble. Pulsed experiments showed the Rabi oscillation frequency dropping from 63 MHz on flat regions to 56 MHz on the pillars, and the spin coherence time falling from about 75 ns to about 58 ns. But the several-fold increase in photon count rate more than compensates, since sensitivity scales with the inverse square root of detected photons. More light means faster data collection and finer magnetic field resolution.
The single-port waveguide architecture contributes its own advantage. Unlike conventional dual-port designs, it absorbs microwave power more efficiently, improving ODMR contrast without requiring higher input power. Simulations confirmed that adding the nanopillar arrays did not disrupt microwave field distribution at the gold stripline, so efficient spin manipulation and plasmonic enhancement operate without interfering with each other.
By uniting plasmonic field concentration, controlled lattice strain, and efficient on-chip microwave delivery in a single device built entirely with standard semiconductor processes, this work opens a practical route toward room-temperature quantum sensors made from two-dimensional materials. Adjusting nanopillar size, spacing, and height offers further design freedom to tune strain and plasmonic response, potentially closing the remaining sensitivity gap between hBN and diamond platforms.
For authors and communications departmentsclick to open
Lay summary
Prefilled posts
Nanowerk Newsletter
Get our Nanotechnology Spotlight updates to your inbox!
Thank you!
You have successfully joined our subscriber list.
Become a Spotlight guest author! Join our large and growing group of guest contributors. Have you just published a scientific paper or have other exciting developments to share with the nanotechnology community? Here is how to publish on nanowerk.com.
