Suppression of Spin-Exchange Relaxation Using Pulsed Parametric Resonance

We demonstrate that spin-exchange dephasing of Larmor precession at near-Earth-scale fields is effectively eliminated by dressing the alkali-metal atom spins in a sequence of ac-coupled $2\pi$ pulses, repeated at the Larmor precession frequency. The contribution of spin-exchange collisions to the spectroscopic linewidth is reduced by a factor of the duty cycle of the pulses. We experimentally demonstrate resonant transverse pumping in magnetic fields as high as 0.1 G, present experimental measurements of the suppressed spin-exchange relaxation, and show enhanced magnetometer response relative to a light-narrowed scalar magnetometer.

Figure 1

Transverse optical pumping with pulsed parametric resonance. A modulating field ${\mathit{B}}_1$ consists of a series of short $2\pi$ pulses separated in time by the dc Larmor precession period $1/\gamma B$. The atoms become polarized along the $\widehat{x}$ direction, and the time-averaged $\widehat{y}$ polarization is sensitive to any differences between the pulse repetition frequency and the dc Larmor frequency.

Figure 2

Fraction of angular momentum lost when a spin-exchange collision occurs during a $2\pi$ magnetic field pulse as a function of $x$ polarization or pulse width $\tau$ (inset). The solid line is the model prediction; the data are results deduced from polarization and relaxation measurements.

Figure 3

Observed $y$ polarization waveforms observed with pulsed parametric modulation. Pulses of the proper repetition rate and area cause rapid $2\pi$ precession of the spins, returning them to he pulses, with no precession between pulses. If the pulse area is incorrect, there is a residual magnetic field between the pulses that causes interpulse precession. If the repetition frequency is off resonance, the atomic polarization is rotated away from the $\widehat{x}$ direction.

Figure 4

Comparison of magnetometer responses for scalar and PPR magnetometers. The main figure shows an optimized experimental comparison at high optical pumping rates, where spin-exchange collisions are partially mitigated for both magnetometers. The magnetic response on resonance is $33\times$ larger for the PPR magnetometer. The inset, taken at low pumping rates, shows directly the reduced effects of spin exchange for the PPR case.