Error-correcting transition pulses for co-located spin ensembles without frequency selectivity

TL;DR

This paper introduces a new class of ultra-fast, robust control pulses for co-located spin ensembles that improve state transfer precision and duration, enabling advancements in fundamental physics tests and quantum memory development.

Contribution

A geometric method for constructing error-correcting, speed-optimized control pulses that do not rely on frequency selectivity and are robust to magnetic field variations and pulse errors.

Findings

Pulses operate at half the quantum speed limit.

Achieved milliradian precision over several hours.

Enabled 30-fold improvement in nuclear-spin state coherence.

Abstract

We present a new class of control pulses designed to transfer co-located ensembles without relying on frequency selectivity, thereby allowing much faster state-transitions. A geometric approach allows us to construct sequences which are robust to changes in the background magnetic field along multiple axes, and errors in the pulse area. \red{These pulses are extremely fast, with robustness to pulse area shown at half the quantum speed limit.} We demonstrate these sequences on nuclear-dipole states, showing milliradian precision over several hours, 30-fold beyond the previous state of the art. This provides a path for extending the coherent integration time of ultra-long-lived nuclear-spin states to the fundamental limit set by their $>$10000 second lifetimes, as the limiting self-interactions of the nuclei are suppressed in the symmetric superposition. The state-preparation quality demonstrated here directly opens up 30-fold improvements in next generation tests of the standard model, especially tests of the symmetries of QCD and searches for dark matter; it is also crucial for the development of nuclear-spin based quantum memories and may be useful in other scenarios demanding extremely fast but robust transitions.