Robust spin relaxometry with fast adaptive Bayesian estimation

TL;DR

This paper introduces an adaptive Bayesian estimation method for spin relaxometry with NV centers, significantly reducing measurement time and increasing robustness against system drifts, enhancing practical applications.

Contribution

The authors develop a fast, adaptive Bayesian approach for NV relaxometry that improves sensitivity and speed, and a four-signal protocol robust to experimental drifts.

Findings

Speedup of relaxometry measurements by an order of magnitude.

Robustness to drifts in contrast, polarization, and pulse fidelity.

Near-optimal sensitivity achieved with the new protocols.

Abstract

Spin relaxometry with nitrogen-vacancy (NV) centers in diamond offers a spectrally selective, atomically localized, and calibrated measurement of microwave-frequency magnetic noise, presenting a versatile probe for condensed matter and biological systems. Typically, relaxation rates are estimated with curve-fitting techniques that do not provide optimal sensitivity, often leading to long acquisition times that are particularly detrimental in systems prone to drift or other dynamics of interest. Here we show that adaptive Bayesian estimation is well suited to this problem, producing dynamic relaxometry pulse sequences that rapidly find an optimal operating regime. In many situations (including the system we employ), this approach can speed the acquisition by an order of magnitude. We also present a four-signal measurement protocol that is robust to drifts in spin readout contrast, polarization, and microwave pulse fidelity while still achieving near-optimal sensitivity. The combined technique offers a practical, hardware-agnostic approach for a wide range of NV relaxometry applications.