Fig. 4.

Scaling of decoupling techniques for improved spectral resolution. a Ramsey oscillations of a different NV with a −1.8 kHz-coupled ¹³C spin (target B ₁) during constant sensor-spin repumping. The decay constant of the oscillation is 23.8 ± 2.9 ms. b Fourier transform of (a). Analogous to Fig. 3c and d, the dark gray points are the transform of the unprocessed measurement data, the light gray line is the transform of the zero-filled signal. The green line is the simultaneous fit of the real and imaginary part of the transformation. The linewidth was determined to be 13.3 ± 1.6 Hz. c Simplified measurement scheme of the target decoherence, whereas the sensor is in its neutral charge state. Before applying the first π/2 pulse on the target spin, the NV center is pumped to the NV⁰ state by applying a 1 ms orange laser pulse with a power of 100 μW. The charge state is afterwards recovered by applying a short green laser pulse. d Simulated sensor limits on coherence times (lines) of a target nuclear spin coupled to an NV center, when using one of the aforementioned decoupling methods. The gray curve is for the plain NV⁻, the red for the NV⁰ case (see c). The orange line shows the coherence time limit, when using the optimal optical repolarization rate for every coupling (see “Methods”). All measured coherence times of this work are displayed as color-coded circles. The shown errors correspond to the standard error of the exponential fit to the decay of the spin state. The gray diamond marks the hypothetical coupling strength at which the plain NV⁻ exerts a dissipative-decoupling effect, which doubles the target spins coherence time limit. In e, the possible location of nuclear spins with said coupling strength of different spin species are shown (see diamond in d)