Nitrogen-vacancy (NV) centers in diamond are atomic-scale magnetometers — single electron spins whose quantum state shifts measurably in response to local magnetic fields. But knowing the field isn't enough. You need to know where the NV center is. For quantum computing (addressing specific qubits) or sensing (localizing magnetic sources), position matters as much as measurement.
Note (2026-11-13; read against the paper's full text): the figures match the abstract (a pulsed field gradient up to 13.5 gauss per micrometre, a single-centre localization of 0.28 plus or minus 0.10 nm, a field measurement deviation of 9 nT), and the paper says the 0.28 nm is the resolution in one dimension. I did not find the lattice spacing of 3.57 angstrom in the text, so the essay's comparison with the diamond lattice is its own; the paper compares instead with confocal microscopy (about 200 nm), stimulated emission depletion (below 10 nm) and the best earlier pixel resolution of about 3.5 nm. The essay's statements that the resolution is a signal-to-noise problem rather than a physics limit, and that the centre encodes its own position and the field simultaneously, are the essay's reading.
The authors (arXiv:2603.22718) achieve 0.28 ± 0.10 nm localization of single NV centers using Fourier magnetic imaging. Pulsed magnetic field gradients up to 13.5 gauss per micrometer encode position as frequency in the NV's response signal. A compact platform with thermal drift compensation keeps the sample stable during measurement.
The spatial resolution is subnanometer — smaller than the diamond lattice spacing (3.57 Å). The magnetic field measurement deviation is 9 nanotesla. Position and field are measured simultaneously from the same quantum response.
The through-claim: the NV center is both the sensor and the thing being sensed. The same spin that measures the external magnetic field also encodes its own position in that field when gradients are applied. The resolution isn't limited by wavelength (optical diffraction) or by the probe size (the NV is a point defect) but by how precisely you can read the frequency-position encoding — which is a signal-to-noise problem, not a physics limit.