Periodic Reporting for period 1 - BoNi-SENS (Quantum Sensing with van der Waals Heterostructures based on hexagonal Boron Nitride)
Okres sprawozdawczy: 2023-02-01 do 2025-07-31
The overall objective of the project is to implement a new quantum sensor embedded in two-dimensional (2D) hexagonal boron nitride (hBN) to study the properties of artificially stacked 2D materials and devices. The essential building block of such van der Waals (vdW) heterostructures is the spin defect in hBN - negatively charged boron vacancy VB– reported by us in 2020. These intrinsic lattice defects can be optically spin-polarized and coherently manipulated, allowing the read-out of quantum information during the spin-coherence time. Although other types of spin centers have been found in hBN since then, this spin-1 center remains the only one with a clearly elucidated structure. Practical applications of hBN with VB- as intrinsic sensors in vdW heterostructures are envisioned, but their implementation lags due to the moderate photoluminescence (PL) quantum yield and spin-coherence times. Furthermore, no vdW heterostructures with enclosed spin-decorated hBN layer have not been reported so far. Our experimental approach is based on coherent manipulation of the spin state using high-frequency pulse protocols, followed by optical readout to explore the adjacent environment, in particular by studying the local lattice strains, pressure, temperature, electric and magnetic fields. Optical readout will be extended by electrical control of spin and charge states, which is an unexplored area and a major step forward in the development of quantum applications of vdW heterostructures.
In the reporting period, we focused on the (i) enhancement of VB- emission and spin resonance contrast by systematically varying the defect density at different irradiation doses.We investigated an hBN flake that was irradiated with a 30 keV nitrogen ion beam in square areas of 4x4 µm with different irradiation fluences ranging from 1.5x 10^11 to 1.1x 10^15 ions/cm². To investigate the sample, a home-built room temperature pulsed ODMR setup equipped with confocal microscope with a single-photon counting module was used together with laser excitation of 473 nm to initialise and read out the defect spin state. We found a correlation between the spin-relaxation times and the fluence of the ion irradiation. The T2* time increased by a factor of two with decreasing fluence. The highest observed T2* was 111 ns, and the data suggest even higher values with less intense irradiation. Considering the brightness as a function of fluence, we find that the generation of VB– with a fluence of 1.5x 10^14 nitrogen ions/cm² gives the best sensitivity in our magnetic field sensor. In summary, although the creation of a single defect has not yet been achieved, we present an irradiation protocol for optimized quantum sensing. For increased irradiation fluence, we observe that PL and thus defect density increases, while spin coherence times decrease, pointing at an optimum intermediate fluence for the desired sensing applications.
Another aspect we focused on was (ii) identifying the sources of spin decoherence of these defects and how to bypass them. The main source of spin decoherence of VB- defect is clearly the magnetic environment, which means that the electronic spin center is surrounded by 100% magnetic nuclei (boron and nitrogen). Without isotopic purification of the crystal, which is probably very expensive if it is realistic at all, it seems difficult to reach the level of spin-coherence times known for carbon-based materials. In order to utilise the main advantages of hBN, namely the two-dimensionality, and avoiding isotope purification of the crystal, we investigated the dynamics of the intermediate state (IS), also called metastable state or shelf state, because it can trap electrons for a certain time. This state plays a significant role in the overall utility of the spin defect and possibly mitigates the requirements for spin coherence times. We investigated the PL dynamics using a simple optical pumping scheme with variable and short delay time between pulses. Our experiments have shown that the IS delays the relaxation of electrons to the ground state with a characteristic lifetime of 24 ns. This lifetime is at least doubled if the temperature is lowered to 4 Kelvin. This has a clear impact on the subsequent sensing protocol and must be considered when designing the pulsed optically detected magnetic resonance (ODMR) experiment. We demonstrate that by doing so, the efficiency of spin manipulation (Rabi oscillation amplitude and ODMR contrast) is increased. We believe that understanding the role of IS in the pump cycle and the degree of spin polarization of the ground state can partially mitigate the requirements of a long spin-spin relaxation time and still ensure the sensitivity of the magnetic field sensor.
Another aspect we focused on was (ii) identifying the sources of spin decoherence of these defects and how to bypass them. The main source of spin decoherence of VB- defect is clearly the magnetic environment, which means that the electronic spin center is surrounded by 100% magnetic nuclei (boron and nitrogen). Without isotopic purification of the crystal, which is probably very expensive if it is realistic at all, it seems difficult to reach the level of spin-coherence times known for carbon-based materials. In order to utilise the main advantages of hBN, namely the two-dimensionality, and avoiding isotope purification of the crystal, we investigated the dynamics of the intermediate state (IS), also called metastable state or shelf state, because it can trap electrons for a certain time. This state plays a significant role in the overall utility of the spin defect and possibly mitigates the requirements for spin coherence times. We investigated the PL dynamics using a simple optical pumping scheme with variable and short delay time between pulses. Our experiments have shown that the IS delays the relaxation of electrons to the ground state with a characteristic lifetime of 24 ns. This lifetime is at least doubled if the temperature is lowered to 4 Kelvin. This has a clear impact on the subsequent sensing protocol and must be considered when designing the pulsed optically detected magnetic resonance (ODMR) experiment. We demonstrate that by doing so, the efficiency of spin manipulation (Rabi oscillation amplitude and ODMR contrast) is increased. We believe that understanding the role of IS in the pump cycle and the degree of spin polarization of the ground state can partially mitigate the requirements of a long spin-spin relaxation time and still ensure the sensitivity of the magnetic field sensor.
We proposed a method for quantifying the spin density in thin hBN layers. Although methods such as electron paramagnetic resonance are available for estimating the spin density, they are hardly suitable for the small absolute number of defects in thin hBN layers. Using widely accessible Raman spectroscopy, we identified new Raman signals, labeled as D1 and D2 alongside the well-established E2g mode in irradiated hBN flakes. We found experimentally and corroborated theoretically (DFT) that these Raman modes are associated with VB- defects, which also exhibit a pronounced PL signal in the near infrared. By analyzing the fluence dependence of these Raman modes along with PL peak, we developed a method to quantify spin density as a function of irradiation dose. This approach was inspired by the challenge of determining defect density in graphene a decade ago, but we have extended the theoretical model to hBN, which also involves PL intensity. We established an empirical relation linking the D1, E2g and PL modes to the irradiation dose. Ultimately, this analysis allows for an easy-to-use all-optical quantification of absolute spin defect density present in hBN and possibly in other 2D materials, making it a universally applicable tool where direct measurement is challenging or impractical.