Revealing and reducing growth-induced interfacial disorder in preferentially aligned nitrogen-vacancy centers in diamond

arXiv:2608.02350 · arXiv: 2608.02350

License: CC BY 4.0.

Abstract

Nitrogen-vacancy (NV) centers in chemical-vapor-deposition (CVD) diamond can form preferentially oriented ensembles with high sensing performance and low densities of lattice defects. Thin films of this material are a cornerstone of various imaging modalities. However, nitrogen injection needed to produce such films can transiently drive growth out of equilibrium, generating interfacial strain and spin defects that degrade NV coherence. Here, we investigate this disorder in $^{12}\text{C}$-enriched, preferentially oriented NV layers grown on (111) diamond using two nitrogen-injection procedures, combined with nanometer-scale selective plasma etching and NV spin-coherence measurements. Pulsed nitrogen injection produces a pronounced nitrogen overshoot within a 60--80 nm interfacial region, generating excessive amounts of defects. By contrast, smooth nitrogen delivery through mass flow controllers substantially suppresses interfacial disorder, yielding coherence properties close to the theoretical limit imposed by spin-bath noise. A 50-nm NV layer is used to demonstrate proton nuclear magnetic resonance detection. This work reveals the role of interfacial disorder associated with the nitrogen-doping procedure and provides a route to growing high-quality, thin NV-doped layers for quantum-sensing applications.

Figures

18 panels with data across 5 figures. Each panel page shows the plot, its columns and its files; each data.csv begins with a header naming the paper, the panel, the source, the license and the provenance route.

Fig. 1

• panel (a): CN/C intensity ratio of the plasma optical emission, on a log axis, against growth time (min) for samples S1 and S2. The red dashed line marks the interface between the buffer layer and the N-doped layer, and the gray dashed line the diamond surface. data.csv
• panel (b): ToF-SIMS nitrogen concentration [N] (ppm) against depth from the surface (nm) for samples S1 and S2. The blue and orange dashed lines mark the interfaces of S1 and S2. data.csv

Fig. 2

• panel (b): Normalized ODMR spectra, Norm. Contrast against $\Delta f$ (MHz), for samples S1 (blue) and S2 (orange): the measured points and the smooth curve the print draws through each. data.csv
• panel (c): Decay rates $\Gamma_2^*$ and $\Gamma_{DQR}/2$ (MHz) for regions R0 to R3 of samples S1 (blue and light blue) and S2 (orange and light orange), with error bars. data.csv

Fig. 3

• panel (a): Experimental (black) and simulated (dark blue) DEER spectra, contrast against drive frequency (MHz). Red dotted lines mark the on-axis and green dotted lines the off-axis P1 transitions. data.csv
• panel (b): DEER decay curves of region R0 of sample S2, contrast against total time ($\mu$s), with the additional $\pi$ pulse off resonance (blue) and on the on-axis transition (orange). Dashed curves are the exponential fits, stretched off resonance and single on axis. The fit residuals of the printed inset are panel b-inset. data.csv
• panel (b-inset): Inset of printed panel (b): fit residuals (contrast) against total time ($\mu$s) for the off-resonance and on-axis curves, about a dashed zero line, each legend entry with the $R^2$ the inset prints. data.csv
• panel (c): DEER decay curves of region R0 of sample S2, contrast against total time ($\mu$s), off resonance (blue) and on the on-axis transition (orange), with their biexponential fits (dotted). The fit residuals of the printed inset are panel c-inset. data.csv
• panel (c-inset): Inset of printed panel (c): residuals of the biexponential fits (contrast) against total time ($\mu$s) for the off-resonance and on-axis curves, about a dashed zero line, each legend entry with the $R^2$ the inset prints. data.csv

Fig. 4

• panel (a-1): [NV] (ppm, left axis, per-layer steps with error bars) and SIMS [N] (ppm, right axis) against depth from the surface (nm) for sample S1. The [P1] profile of printed panel (a) is panel a-2, and the gray dotted layer boundaries of the print are not drawn. data.csv
• panel (a-2): [P1] (ppm) against depth from the surface (nm) for sample S1, as per-layer steps with error bars: the third concentration scale of printed panel (a). The [NV] and [N] profiles are panel a-1. data.csv
• panel (b-1): [NV] (ppm, left axis, per-layer steps with error bars) and SIMS [N] (ppm, right axis) against depth from the surface (nm) for sample S2. The [P1] profile of printed panel (b) is panel b-2, and the gray dotted layer boundaries of the print are not drawn. data.csv
• panel (b-2): [P1] (ppm) against depth from the surface (nm) for sample S2, as per-layer steps with error bars: the third concentration scale of printed panel (b). The [NV] and [N] profiles are panel b-1. data.csv
• panel (c): Conversion ratios per layer L0 to L3 for sample S1: P1/N (%, blue, left axis) and NV/P1 (%, orange, right axis), with error bars. data.csv
• panel (d): Conversion ratios per layer L0 to L3 for sample S2: P1/N (%, blue, left axis) and NV/P1 (%, orange, right axis), with error bars. data.csv
• panel (e): Spin-echo decay rate $\Gamma_2$ (kHz) per layer L0 to L3 for sample S1: measured, and projected from the P1-center concentrations and from the SIMS nitrogen concentrations, with error bars. data.csv
• panel (f): Spin-echo decay rate $\Gamma_2$ (kHz) per layer L0 to L3 for sample S2: measured, and projected from the P1-center concentrations and from the SIMS nitrogen concentrations, with error bars. data.csv

Fig. 5

• panel (1): $^{1}$H NMR spectra of region R2 of sample S1, contrast against DD pulse interval ($\mu$s), for KDD8 (blue) and KDD16 (orange), 8 and 16 repetitions of the KDD sequence. The dashed curves are Lorentzian fits to the two spectra; the print draws its fits dotted. data.csv

Cite

Cheng-I Ho, Marina Davydova, Patrik Straňák, Felix Hoffmann, Peter Knittel, Andrej Denisenko, Jörg Wrachtrup. Revealing and reducing growth-induced interfacial disorder in preferentially aligned nitrogen-vacancy centers in diamond. arXiv:2608.02350

When you use hosted data, cite the original paper and give the panel's URL so a reader can find the exact values you used.

All papers · About · Statistics · Submit data · API