Negatively charged nitrogen-vacancy (NV−) centers in diamond are extensively recognized as exceptional nanoscale solid-state spin quantum systems [1–4]. These centers exhibit long spin-coherence times and high-efficiency optical transitions at room temperature, enabling convenient optical initialization and readout of quantum states [5–8]. The Hamiltonian of diamond NV color centers reveals significant potential for detecting various physical quantities, including magnetic fields, stress, temperature, and electric fields [9,10]. Despite these advantages, the fluorescence collection efficiency of diamond NV color-center ensembles is typically low. Moreover, the electron and nuclear spins from impurity nitrogen atoms significantly interfere with the NV centers, thereby limiting their sensitivity as quantum sensors [11].

Compared with polycrystalline diamond and nanodiamond, NV color centers in single-crystal diamonds demonstrate superior ground-state lifetimes and quantum properties [12]. Recent research has identified two primary approaches to enhancing the sensitivity and fluorescence intensity of single-crystal diamond NV color centers: reduced-material manufacturing and additive manufacturing. Reduced-material manufacturing includes techniques such as reactive-ion etching, focused ion beam etching, inductively coupled plasma etching, and electron beam lithography, which transform single-crystal diamonds into two-dimensional photonic crystals with air columns or diamond micro-columns [13–15]. Additive manufacturing, often referred to as template-assisted growth, involves growing diamond photonic crystals with anti-opal or other structures using various masks [16,17]. Although these methods can yield diamond microcavities with excellent optical properties, they often involve complex micro-nano machining processes and are associated with high costs.

One-dimensional (1D) photonic crystals typically consist of two materials with different permittivities arranged periodically [18–20]. These structures can suppress the propagation of electromagnetic waves in a specific direction by manipulating the refractive index and thickness of the dielectric materials. In this study, we deposited a 1D photonic crystal structure on the lower surface of single crystal diamond using a periodic combination of high refractive index dielectric film titanium dioxide (TiO2, n=2.61) and low-refractive-index silica (SiO2, n=1.45) [21]. The fluorescence wavelength of the NV color center is in the photonic band gap by optimizing the thickness of TiO2 and SiO2 dielectric films. This reduces the fluorescence of the NV color center emitted from the lower surface of the diamond when excited by a pump laser. In addition, the reflected pump laser can excite more NV color centers, thereby improving the excitation efficiency and fluorescence collection efficiency of NV color centers. Figure 1 depicts a schematic of the fluorescence emitted by the diamond NV color centers upon excitation by a pump laser.

Furthermore, the electron-spin interference is mitigated as the reflected laser excites the localized spin electrons of the impurity nitrogen atoms into a state of synchronous motion. The optical-detection magnetic resonance system measurements and theoretical calculations demonstrate that this approach significantly enhances the sensitivity of the diamond NV color-center ensemble as a magnetometer.

Two chemical-vapor-deposited (CVD) single-crystal diamonds (the concentration of the NV color center is about 0.8 ppb; ppb, parts per billion), each measuring 5mm×5mm×0.3mm, were selected for this study as sample 1 and sample 2. Initially, the samples were cleaned with anhydrous ethanol, acetone, and deionized water for 20 min each. They were then subjected to a heating process at 250°C for 2 h in a mixed acid solution of HNO3 and H2SO4 in a 1:1 ratio [27]. According to the simulation results of FDTD in Fig. 2(b), it is shown that when the number of periods of TiO2 and SiO2 alternating dielectric film is six, the reflectivity at the ZPL of NV− color center has almost reached the maximum. Therefore, six periods of TiO2 and SiO2 alternating dielectric films were deposited on the surfaces of both diamond samples using magnetron sputtering.

The transmittance of the samples within the NV color-center fluorescence band (550–700 nm) was measured using a dual-beam ultraviolet-visible (UV-VIS) photometer before and after deposition. The cross-sections of the alternating TiO2 and SiO2 films were characterized using scanning electron microscopy (SEM). The photoluminescence (PL) and Raman spectra were obtained using a Horiba HR Evolution Raman spectrometer, operating with an excitation light power of 10 mW, a grating of 1800 g/mm, a confocal aperture of 50 μm, a 50× objective lens, and a numerical aperture of 0.5. X-ray diffraction (XRD) was used to investigate the crystalline orientations of the deposited films. Additionally, the response of the NV centers to magnetic fields and their spin-coherence properties were characterized using an optically detected magnetic resonance (ODMR) system [28,29]. The system utilized a 10 mW laser to irradiate the diamond to initialize the spin state of the NV color center electrons and then act on the NV color center with a continuous microwave intensity of 1 dBm. The microwave frequency is swept in the range of 2750–2950 MHz, the detection time is 200 ms, and finally the ODMR spectrum of the diamond NV color center is read out under the condition of no external magnetism.

Figure 3(a) displays an SEM image of a 1D photonic crystal with a periodic structure fabricated on the surface of sample 2. In the image, the bright fringes indicate SiO2, while the dark fringes indicate TiO2. After annealing sample 2 at 700°C for 1 h in a vacuum, XRD analysis was performed to further investigate the crystalline characteristics of the deposited film, as illustrated in Fig. 3(b). The prominent diffraction peaks at 25.3°, 37.8°, 48.0°, and 54.0° correspond to the (101), (004), (200), and (105) crystal planes of anatase TiO2, respectively. Figure 3(c) shows a noticeable peak at 20°, indicating the formation of amorphous SiO2.

Figure 4(a) shows the transmittance curves before and after the deposition of 1D photonic crystals (TiO2/SiO2 composite periodic film structure) on the surface of diamond sample 1 and sample 2. The high reflectivity presented after the preparation of 1D photonic crystals on the surface of the diamond is the photonic band gap, and its width is about 150 nm. The photonic band gap center wavelength of diamond sample 1 is 637 nm (NV− color center ZPL), and the ZPL of the NV0 color center (575 nm) and NV− color center is in the photonic band gap, while the photonic band gap center wavelength of diamond sample 2 is 697 nm. Such optimization makes only the ZPL of NV− in the photonic band gap, because only the fluorescence intensity of the NV− color center is beneficial for magnetic field measurement [30].

The results illustrated in Fig. 4(b) indicate that a central photonic bandgap wavelength of 637 nm enhanced the fluorescence intensities of both NV0 and NV− color centers owing to the wide bandgap. However, Fig. 4(c) reveals that when the central wavelength was adjusted to 697 nm, the fluorescence intensity of the NV− center increased drastically, whereas that of the NV0 center remained almost unchanged or decreased. In addition, according to Fig. 4(a), it can be seen that the band gap of the photonic crystal on the surface of sample 1 is in the band of 570–720 nm, while that of the photonic crystal on the surface of sample 2 is in the band of 620–770 nm. Since the fluorescence of the diamond NV color center is mainly in the band of 600–800 nm, when the central wavelength of the photonic crystal band gap is 697 nm, the fluorescence of the NV color center can be significantly improved, as shown in Fig. 4(c).

To further investigate the influence of 1D photonic crystals composed of TiO2 and SiO2 on the fluorescence intensity of diamond NV color centers, PL integral mapping was performed by scanning the ZPL and phonon sideband (573–578 nm) for NV0 centers and the ZPL and phonon sideband (636–640 nm) for NV− centers in sample 2, both before and after deposition. As illustrated in Fig. 5(a), the fluorescence intensity of NV0 color centers showed minimal change post-treatment. However, Fig. 5(b) demonstrates a significant enhancement in the fluorescence intensity of the NV− color centers after treatment. Figures 5(c) and 5(d) present the relative fluorescence frequency distributions of NV0 and NV− centers before and after the fabrication of 1D photonic crystals on the surface of sample 2. Thus, 1D photonic crystals enhance the fluorescence intensity of NV− color centers and lead to a narrow relative frequency distribution of these centers.

The quantum coherence properties of the NV color centers in sample 2, both before and after deposition, were measured using an ODMR system. As shown in Fig. 6(a), the ODMR spectra fitted with Lorentzian lines revealed that the full width at half maximum (FWHM) of the lines decreased from 30.6 to 16.4 MHz after treatment, with a noticeable improvement in contrast. Figure 6(b) indicates that the dephasing time, T2*, measured using the Ramsey sequence, increased from 209 to 841 ns. Moreover, according to the results of the improvement of fluorescence collection efficiency and Eq. (1) [28,31], the calculation results show that the magnetic sensitivity of sample 2 is increased by a factor of 2.92 after 1D photonic crystal is fabricated on the surface of diamond, ηCW=433hgeμBΔνCcwR,(1)where h is the Planck constant, ge≈2.003 is the electronic g-factor of the NV− center, μB is the Bohr magneton, R is the photon detection rate, Δν is the linewidth, and Ccw represents the continuous wave ODMR contrast. Furthermore, the concentration of nitrogen impurities in CVD sample 2 was determined using optical absorption spectroscopy [32]. Figure 6(c) presents the UV-VIS apparent absorption spectra of sample 2, with the inset displaying the transmission of the diamond in the UV-VIS range. The measured nitrogen impurity concentration in sample 2 was 2.223 ppm (parts per million), whereas the ODMR measurement results indicate that the concentration of the NV color-center ensembles was 0.783 ppb. Such high concentrations of nitrogen impurities severely affect the quantum sensing capabilities of NV ensembles.

According to Fig. 4(a), 1D photonic crystal also reflects the incident laser (532 nm) to a certain extent, which allows more photons to interact with the NV color centers, improving the fluorescence collection efficiency and signal-to-noise ratio of the NV color centers, thus resulting in a decrease in the FWHM of the ODMR spectrum [33]. In addition, the reflected laser excites the impurity electrons around the NV, improves the electric field environment around the NV color centers (as shown in Fig. 7), and reduces the coupling effect of the impurity electrons with the electron spin of the NV color center [34], thus resulting in the prolongation of the dephasing time, T2*.

In order to verify that the 1D photonic crystal structure equivalently enhances the power of the pump laser, we measured the ODMR spectra of sample 3 (the size and other parameters were similar to those of sample 2 before treatment, and the concentration of the NV color center was 0.8 ppb) with different laser power (mW levels) polarized NV color center spin states, as shown in Fig. 8(a). We used the Gaussian function to simulate the splitting peak of the ODMR spectrum and obtained its FWHM. The dependence of fluorescence intensity and FWHM of the ODMR spectra on laser power is shown in Fig. 8(b), and as the laser power increases, the fluorescence of the collected NV color center increases, and the FWHM of the ODMR spectrum gradually decreases, indicating that our equivalence is correct.

1D photonic crystals on the surface of single-crystal diamonds were successfully fabricated in this study, effectively regulating the direction and significantly enhancing the fluorescence intensity of the diamond NV color-center ensembles. Furthermore, the dephasing time, T2*, increased from 209 to 841 ns, and the sensitivity of the NV color centers as a magnetometer improved by a factor of 2.92. This enhancement underscores the potential of this approach to significantly enhance the sensitivity of diamond NV color-center ensembles for quantum sensing applications. Unlike traditional methods that focus on fabricating photonic crystal microcavities to enhance fluorescence, the proposed approach offers a novel strategy for improving the performance of diamond NV color-center ensembles. This innovation paves the way for the development of high-sensitivity diamond-based quantum devices, marking a significant advancement in the field of quantum sensing technology.