High‑Density Nitrogen‑Vacancy Centers in 5 nm Detonation Nanodiamonds: Photoluminescence, Magnetic‑Field Sensitivity, and Quantitative EPR Analysis

Abstract

Detonation nanodiamonds (DNDs) of <5 nm diameter were shown to host nitrogen‑vacancy (NV⁻) color centers at a concentration of 1.1 ± 0.3 ppm, a record for nanodiamonds with intentionally created NV⁻ centers. The concentration was derived from electron paramagnetic resonance (EPR) using the integrated intensity of the g = 4.27 half‑field line, which originates from forbidden Δm = 2 transitions in the NV⁻ ground‑state triplet. Confocal fluorescence microscopy revealed bright red photoluminescence (PL) from individual NV⁻ emitters, and a pronounced PL drop was observed when the ground‑ and excited‑state spin levels crossed at an external magnetic field of ≈ 100 mT (ground‑state level anti‑crossing, GSLAC). These results demonstrate the quantum‑sensitive nature of NV⁻ centers in sub‑10 nm DNDs and open new avenues for high‑resolution sensing and imaging applications.

Background

NV⁻ centers in diamond are unique quantum defects that combine photostability, room‑temperature spin coherence, and optically detectable magnetic resonance (ODMR). These properties enable applications ranging from nanoscale magnetometry to bio‑labeling and quantum information processing. However, conventional fluorescent nanodiamonds (FNDs) are typically produced by high‑pressure high‑temperature (HPHT) synthesis and contain limited NV⁻ densities, while detonation nanodiamonds (DNDs) offer orders‑of‑magnitude higher surface‑to‑volume ratios but suffer from high defect concentrations that can quench NV⁻ fluorescence. Understanding and quantifying NV⁻ populations in DNDs is therefore essential for realizing their full potential.

Experimental

Detonation Synthesis and Purification

≈ 1 kg of a trinitrotoluene‑hexogen explosive mixture was detonated in a water shell inside a stainless‑steel vessel. The rapid (≈ 10⁻⁵ s) decomposition produced 5 nm diamond crystallites that were subsequently purified in boiling acid to remove 3d transition metals, then annealed in air at 430 °C for 10–12 h to eliminate sp²‑carbon and other surface contaminants. The resulting powder exhibited coherent scattering region (CSR) sizes of 4.5–5.7 nm, with a representative sample of 5.2 ± 0.2 nm used for all major measurements.

EPR, XPS, and Optical Characterization

Room‑temperature X‑band EPR (≈ 9.4 GHz) was performed on 10 mg DNDs in quartz tubes. The half‑field region (130–180 mT) displayed two g‑lines: g = 4.27 (NV⁻) and g = 4.00 (multivacancies). Integration of the g = 4.27 line, after background subtraction, yielded an NV⁻ concentration of 1.1 ± 0.3 ppm relative to a HPHT reference with 5.3 ppm NV⁻. X‑ray photoelectron spectroscopy (XPS) on Ar‑ion‑etched samples revealed an interior nitrogen content of 1.65 ± 0.05 at.% mainly in sp³‑bonded clusters; isolated paramagnetic N s⁰ atoms constituted < 10 % of total nitrogen.

Photoluminescence (PL) spectra were recorded with a 532 nm laser (0.5 mW) and a 100× NA = 1.40 objective, covering 540–1000 nm. The PL maximum at 680 nm is characteristic of NV⁻/NV⁰ ensembles. Confocal microscopy on spin‑coated DND aggregates (average aggregate size 30 nm) produced 2‑D maps of individual emitters. Under a permanent magnet (≈ 90–100 mT), the PL intensity of isolated aggregates dropped sharply, confirming magnetic‑field sensitivity of NV⁻ centers at the GSLAC.

Results and Discussion

Quantitative NV⁻ Counting by Half‑Field EPR

The g = 4.27 line corresponds to forbidden Δm = 2 transitions of the NV⁻ ground‑state triplet. Its integrated intensity, after correcting for overlapping g = 4.00 multivacancy signals, is 4.8 × lower than that of the HPHT reference, giving 1.1 ± 0.3 ppm NV⁻. This value matches the theoretical probability calculated from the average numbers of substitutional nitrogen (≈ 7–8) and vacancies (≈ 7–8) per 5‑nm particle, supporting the random incorporation model during detonation.

Relationship Between Nitrogen Content and PL Intensity

Across a series of DND samples with CSR sizes 4.3–5.6 nm, nitrogen content decreased with increasing CSR, while PL intensity at 680 nm increased. The trend indicates that excess nitrogen in the form of A‑centers (N–N dimers) acts as non‑radiative quenchers of NV⁻ fluorescence. Reducing nitrogen impurities—either by detonation additives or post‑synthetic high‑pressure, high‑temperature treatments—could therefore enhance NV⁻ brightness.

Magnetic‑Field‑Induced PL Modulation

Applying a magnetic field of ≈ 100 mT mixes the |0⟩ and |–1⟩ ground‑state spin sublevels at the GSLAC, redistributing population into the non‑radiative |–1⟩ channel and reducing PL by up to 30 %. This effect, observed in single DND aggregates, is markedly stronger than in HPHT micro‑diamonds and can be exploited for background‑free detection of sub‑diffraction emitters in biological or high‑autofluorescence environments.

Sub‑Diffraction NV⁻ Emitters

Confocal maps revealed isolated bright spots with lateral dimensions below 70 nm, comparable to the diffraction limit (≈ 230 nm for 532 nm excitation, NA = 1.40). Each ~30 nm aggregate contains ~1.3 NV⁻ centers, sufficient for single‑emitter detection. Such sub‑diffraction NV⁻ sources are promising for high‑resolution imaging and quantum sensing in complex media.

Conclusion

5 nm DNDs synthesized by detonation host NV⁻ centers at 1.1 ± 0.3 ppm, a density five‑fold higher than typical HPHT FNDs of 100 nm. NV⁻ PL is clearly observable after removal of surface sp²‑carbon, and the emission is highly responsive to external magnetic fields via the GSLAC. The combined EPR/XPS/PL approach provides a reliable route to quantify and optimize NV⁻ populations in DNDs, paving the way for ultrafine, magnetically tunable quantum emitters in biomedical imaging and nanoscale sensing.