Herein, a microfabricated millimeter-level vapor alkali cell with a high hermeticity is fabricated through a wet etching and single-chip anodic bonding process. The vapor cell, containing Rb and N₂, was investigated in a coherent population trapping (CPT) setup for the application of a chip-scale atomic clock (CSAC). The contrast of CPT resonance is up to 1.1% within the only 1 mm length of light interacting with atom. The effects of some critical external parameters on the CPT resonance, such as laser intensity, cell temperature, and buffer gas pressure, are thoroughly studied and optimized. The improved microfabricated vapor cell also exhibited great potential for other chip-scale atomic devices.

We experimentally investigated continuously frequency-tunable microwave (MW) magnetic field detection based on Rabi resonance with a single cesium-rubidium hybrid vapor cell. The multispecies atomic systems, with their tunable abilities in transition frequencies, enabled this atomic sensing head to cover a broader detectable MW field scope compared to the use of a single alkali atom. Here, we demonstrated the simultaneous observation of atomic Rabi resonance signals with ⁸⁵Rb, ⁸⁷Rb, and ¹³³Cs in the same vapor cell. Using an experimentally feasible static magnetic field (DC field) below 500 G, we realized a MW magnetic field strength detection with a bandwidth of 4.8 GHz around 8.1 GHz. The use of a hybrid cell system cell also enabled the detection of an identical MW field for different atomic species with the help of the DC field, allowing us to perform a perfect comparison for different applications that require the same electromagnetic environment. The results may be useful for the realization and application of many atomic detectors based on different physical principles.

We have proposed a new technique to obtain highly polarized atoms for improving the sensitivity of the all-optical Cs atomic magnetometer. In a Bell–Bloom magnetometer, the resonance pumping from the ground state F g = 3 to the excited state F e = 4 at the D1 line is used to polarize the Cs atoms and populate the sublevels of the | F g = 4 , m F = 4 ⟩ partly as the transition F g = 4 → F e = 3 remains leaking. Here we utilize a modulation technique to generate a sideband frequency to pump both transitions simultaneously, which drives more atoms to the target state | F g = 4 , m F = 4 ⟩ . This improvement significantly increases the magnetic resonance signal but with little effect on its linewidth. It is very suitable for the miniaturization of the atomic magnetometer in the future.

This paper reports an all-optical vector magnetometer enhanced by a machine learning model. Using a dual probing beam setup, spin projections in both probe directions are simultaneously detected. Vector information is directly obtained from the measured phases of spin projection signals. To enhance the measurement accuracy and mitigate the dead zone effect, we introduce an artificial neural network (ANN) to link the phase signals to the field direction. With the addition of amplitude information to the ANN input, the average angle error is reduced to less than 0.3° within a hemisphere. Furthermore, this configuration demonstrates a field angle sensitivity of better than 30 μ rad/Hz1/2.