Reducing frequency drift caused by light shift in coherent population trapping-based low-power atomic clocks

Abstract

We propose a model to reduce the influence that the light-shift induced frequency drift has on the long-term stability of coherent population trapping (CPT)-based low-power atomic clocks. We experimentally validated the proposed model using the chip-scale atomic clock architecture. The model considers both the drift of the dc-bias current to compensate for the wavelength aging of the vertical-cavity surface-emitting laser (VCSEL) and the variation in the modulation index of the light field generated by the VCSEL. We investigated the dc-bias-current-dependency of the frequency drift in an atomic clock module both theoretically and experimentally, confirming that the frequency drift can change as a function of the VCSEL basic parameters. When the atomic clock module was operated over a period of five months at the dc-bias current affording a zero-crossing of the clock frequency drift, the Allan standard deviation and clock frequency drift for long-term stability were 8.0×10−12 for averaging times of 4×106 s and 2.4×10−13/day, respectively. Our approach is promising to improve the long-term stability of CPT-based low-power atomic clocks.

Short-term stability of Cs microcell-stabilized lasers using dual-frequency sub-Doppler spectroscopy

“…The interrogation of an alkali vapor atom ensemble in a mmscale cell has allowed the development of a wide variety of high-precision chip-scale atomic devices [1]. Among these instruments, miniaturized microwave atomic clocks based on coherent population trapping (CPT) [2] have known a remarkable development and progress, including their commercialization [3], due to their unrivaled size-power-frequency stability budget [4][5][6][7]. These clocks are now widely-used in numerous applications, including underwater sensor networks, secure communications or satellite-based navigation systems.…”

“…However, microwave CSACs suffer from drawbacks and limitations. These clocks use in general vertical-cavity surfaceemitting lasers (VCSELs) whose frequency noise and variations can limit the clock frequency stability [7,8]. In addition, the presence of buffer gas pressure in the cell induces an important frequency shift of the clock transition [9] that can limit the clock mid-term frequency stability and jeopardize the establishment of a clock accuracy budget.…”

Advanced light-shift compensation protocol in a continuous-wave microcell atomic clock

“…In optically-pumped Rb clocks, a method of suppressing the light-shift by adjusting the laser frequency used for optical pumping was proposed in [35]. In CPT atomic clocks, several approaches for light-shift mitigation have been proposed, including the fine tuning of the VCSEL microwave modulation index [36,37], the compensation for the laser frequency detuning [38], the adjustment of the cell temperature to a specific setpoint [39] or solutions combining actions on both the VCSEL dc-bias current drift and on the modu-lation index of the VCSEL light field [40]. Nevertheless, the latter techniques rely on experimentally-determined quantities that are expected to change with time and from one device to another.…”

Light-shift mitigation in a microcell-based atomic clock with Symmetric Auto-Balanced Ramsey spectroscopy

“…Some efforts were oriented towards the extraction of the actual laser or cell temperature, from the atomic response itself, in order to reduce the negative impact of temperature gradients between these key components and their respective temperature sensors 2,5,7 . Other sophisticated methods include the active stabilization of a specific laser microwave modulation index that reduces laser power-induced frequency instabilities 3,5,[8][9][10] , possibly combined with the compensation for the laser aging 6 , or the implementation of advanced tailored interrogation sequences using laser power modulation 11 . Deposition of gold micro-discs, used as privileged alkali condensation spots onto the cell windows was also shown to avoid the progressive obstruction of the transmitted laser light 12 .…”

“…These values are high in comparison with those reported in the literature 3,28,32 . Variations ∆P µW of the microwave power of only 0.06 dB at 1 day might be more suspected 3,6,28 to justify the stability limitation. However, this assumption is counterbalanced by two observations.…”