Operating an atom-interferometry-based gravity gradiometer by the dual-fringe-locking method

We report a gravity gradiometer composed of two atom interferometers which are simultaneously locked to the respective fringe center by feedback control of the Raman laser phase and magnetic field gradient. Compared with the conventional full-fringe recording method, this dual-fringe-locking strategy is capable of increasing the sampling rate and improving the sensitivity for gravity gradient measurements. Meanwhile, it retains the intrinsic advantage of rejecting common mode noises for gravity gradiometers. A short-term sensitivity of $670 \mathrm{E}/{\mathrm{Hz}}^{1/2}$ with a 0.25 Hz sampling rate is achieved in our gravity gradiometer. This approach may also find possible applications in other atom-interferometry-based experiments utilizing dual atomic clouds.

Figure 1

Schematic of the experimental setup. The two atomic clouds are freely falling with almost the same velocity at different heights. The feedback coil used for locking fringe $I$ is close to the apogee of cloud $I$, while the test coil used to induce an additional phase shift for the test is close to the apogee of cloud $\mathit{II}$.

Figure 2

The induced phase shift by the feedback coil at different values of injected current, with (a) for cloud $I$ and (b) for cloud $\mathit{II}$.

Figure 3

The switch between full-fringe recording mode and fringe-locking mode, with (a) for cloud $I$ and (b) for cloud $\mathit{II}$, respectively. The solid line in each picture is a sinusoidal fit of the experimental points for each cloud.

Figure 4

The performances characterized by Allan deviation. The Allan deviation of the gravity gradient measurement in the full-fringe recording mode is displayed as open circles, and the one in the fringe-locking mode is displayed as solid squares.

Figure 5

The measured variation between the cases of taking masses on and off by our gravity gradiometer. The two dashed lines indicate the theoretical variation.

Figure 6

The result of the validation experiment. The measured phase shift is induced by the current of the feedback coil according to Eq. (4).