Towards accurate vector magnetometry with NV centers in diamond - Lilian Childress

Over the last two decades, research into magnetometry with nitrogen-vacancy (NV) centers in diamond has largely ignored the question of accuracy, possibly because it was deemed uninteresting, or possibly because high accuracy was considered unachievable. Nevertheless, in practical applications such as geomagnetic mapping, accuracy can be as or even more important than sensitivity. Startup SBQuantum has recently demonstrated commercially competitive vector accuracy with NV magnetometers, but the physics behind their machine-learning calibration and the ultimate limits to accuracy in NV magnetometry remain open questions. This talk will examine accuracy in NV magnetometry from two perspectives: eliminating drift-prone bias fields and identifying a significant source of systematic error. The bias magnetic field commonly applied to NV ensembles to isolate signals from different defect orientations introduces a source of drift indistinguishable from the external field to be measured. Using a pulsed 2D spectroscopy technique, we show it is possible to separate signals from different orientations even when their spin transition frequencies overlap, opening the door to bias-field-free operation. However, removing the bias magnetic field augments the effects of internal electric fields coming from charged dopants within the diamond. Numerical simulations indicate that these internal electric fields can introduce significant systematic errors for dense NV ensembles, exceeding other known sources of uncertainty even in ~mT bias fields. Our results thus highlight some of the opportunities and challenges in understanding and improving the accuracy of NV magnetometry.
This work was done at and/or in collaboration with SBQuantum.