Suppression of heading error induced by vector light shift in a CPT atomic magnetometer using a double-pass configuration

Coherent Population Trapping (CPT) atomic magnetometers hold broad application prospects in fields such as geological exploration, earthquake monitoring and early warning, marine magnetic anomaly detection, and space magnetic field mapping due to their significant advantages of all-optical configuration, omnidirectional detection, and ease of miniaturization and integration. However, when deployed on mobile platforms, the accuracy of CPT magnetometers is strongly constrained by the heading error. In the present work, we identify the vector light shift (VLS) as a key contributor to this heading error through theoretical analysis and experimental validation. To mitigate this effect, a VLS suppression method based on a double-pass configuration is proposed and experimentally validated. Compared to the conventional single-pass setup, this approach reduces the fictitious magnetic field corresponding to the VLS from 7.58 nT to 4.61 nT, achieving a suppression ratio of 39.2%. This method effectively mitigates the impact of the VLS on the systematic error and heading error of the CPT magnetometer. Furthermore, the experiment reveals that this method also enhances the measurement stability of the CPT magnetometer. At an integration time of 6.6 s, a 45.2% reduction is observed in the Allan deviation of the magnetic field reading. This study provides an effective heading error suppression scheme for the development of high precision CPT magnetometers for mobile platforms.