Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

VCSELs for Atomic Sensors.

Darwin K. Serkland

Published Jul 26, 2016• 0 citations
OpenAlex

Abstract

A new generation of small low-power atomic sensors, including clocks, magnetometers, and gyroscopes, is being developed based on recently available MEMS and VCSEL technologies.These sensors rely on spectroscopic interrogation of alkali atoms, typically rubidium or cesium, contained in small vapor cells.The relevant spectroscopic wavelengths (in vacuum) are 894.6 nm (D1) and 852.3 nm (D2) for cesium, and 795.0 nm (D1) and 780.2 nm (D2) for rubidium.The D1 wavelengths are either preferred or required, depending on the application, and vertical-cavity surface-emitting lasers (VCSELs) are preferred optical sources because of their low power consumption and circular output beam.This paper describes the required VCSEL characteristics for atomic clocks and magnetometers.The fundamental VCSEL requirement is single-frequency output with tunability to the particular spectroscopic line of interest.Singlepolarization and single-transverse-mode operation are implicit requirements.VCSEL amplitude noise and frequency noise are also important because they contribute significantly to the sensor signal-to-noise ratio.Additional desired VCSEL attributes are low cost, low power consumption, and several years of continuous operating lifetime.This paper also describes the 894-nm VCSELs that we have developed for cesium-based atomic sensors.In particular, we discuss VCSEL noise measurements and accelerated lifetime testing.Finally, we report the performance of prototype atomic clocks employing VCSELs.

Related papers

VCSELs for atomic clocks

Darwin K. Serkland, Gregory M. Peake, K.M. Geib +4 more

The spectroscopic technique of coherent population trapping (CPT) enables an all-optical interrogation of the groundstate hyperfine splitting of cesium (or rubidium), compared to the optical-microwave double resonance technique conventionally employed in atomic frequency standards. All-optical interrogation enables the reduction of the size and power consumption of an atomic clock by two orders of magnitude, and vertical-cavity surface-emitting lasers (VCSELs) are preferred optical sources due to their low power consumption and circular output beam. Several research teams are currently using VCSELs for DARPA's chip-scale atomic clock (CSAC) program with the goal of producing an atomic clock having a volume < 1 cm^3, a power consumption < 30 mW, and an instability (Allan deviation) < 1x10^-11 during a 1-hour averaging interval. This paper describes the VCSEL requirements for CPT-based atomic clocks, which include single mode operation, single polarization operation, modulation bandwidth > 4 GHz, low power consumption (for the CSAC), narrow linewidth, and low relative intensity noise (RIN). A significant manufacturing challenge is to reproducibly obtain the required wavelength at the specified VCSEL operating temperature and drive current. Data are presented that show the advantage of operating at the D1 (rather than D2) resonance of the alkali atoms. Measurements of VCSEL linewidth will be discussed in particular, since atomic clock performance is especially sensitive to this parameter.

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE•Feb 9, 2006•60 citations

Reducing PM-to-AM Noise Conversion in Atomic Vapors

Kaitlin Harpenau, Nathaniel Ristoff, James C. Camparo

Laser phase noise (PM) to transmitted intensity noise (AM) conversion is a fundamental aspect of the field/matter interaction, and a major noise process that limits the frequency stability of next-generation atomic clocks. It arises from the absorption cross section’s dependence on laser phase and is especially pernicious for vertical-cavity surface-emitting lasers (VCSELs) given their relatively broad singlemode linewidths (e.g., 40 MHz). This is unfortunate, since VCSELs (with their very low operating power and low probability for mode hops) are otherwise ideal for quantum devices. Here, we demonstrate that atomic spectroscopy with N VCSELs reduces the total PM-to-AM noise by a factor of ~N1/2. Consequently, atomic clocks with (for example) five VCSELs would have a signal-to-noise ratio (SNR5) approximately 120% higher than the same device operating with one VCSEL (SNR1) at the same total power: SNR5/SNR1≌ 2.2.

IEEE Open Journal of Ultrasonics Ferroelectrics and Frequency Control•Jan 1, 2026•0 citations

The Chip-Scale Atomic Clock - Recent Development Progress

Robert Lutwak, Donald A. Emmons, Thomas C. English +1 more

We have undertaken the development of a chip-scale atomic clock (CSAC) whose design goals include short-term stability, sigma(sub y)(tau=1 hour), of 1x10(exp 11) with a total power consumption of 30 mW and an overall device volume of 1 cu cm. The stringent power requirement dominates the physics package architecture, dictating a small (<5 mm(exp 3) gaseous atomic ensemble interrogated by a low-power semiconductor laser. At PTTI 2002, we reported on initial experimental investigations leading to the decision to employ the coherent population trapping (CPT) interrogation technique. This paper describes our further progress on the CSAC effort, including the development of custom vertical cavity surface emitting laser (VCSEL) sources and techniques for microfabricating miniature cesium vapor cells comprised of anodically bonded silicon and glass. Measurements of the signal contrast and linewidth of both the cesium D1 and D2 resonance transitions are compared, and frequency stability measurements of the CSAC testbed are presented.

•Sep 1, 2004•55 citations

System-level evaluation of 795nm Q-VCSELs in 87Rb qSensors

Amirhossein Ghods, Newton Rai, G A Garcia +3 more

We report system-level evaluation of manufacturable, quantum-optimized 795-nm vertical-cavity surface-emitting lasers (Q-VCSELs) for rubidium (Rb) atomic sensors, with emphasis on integration into an 87Rb coherent population trapping (CPT) clock prototype. Operation at 795 nm (Rb D1) is constrained by increased internal loss/absorption, polarization instability, and stringent requirements on spectral purity. At the component level, we employ epitaxial and layout engineering to improve mode discrimination, polarization robustness, and high-temperature efficiency, achieving stable single-transverse-mode emission and linear polarization from 20–80 °C with side-mode suppression ratio (SMSR) >30 dB and polarization extinction ratio (PER) >20 dB. The devices exhibit low threshold current (<400 µA at 80 °C) and slope efficiency >0.5 mW/mA, enabling >1 mW single-mode, single-polarization output power in chip-scale form factors. Quantum-relevant characterization shows linewidth on the order of ~30–35 MHz and relative intensity noise below −130 dB/Hz at frequencies greater than 1KHz. We then integrate the Q-VCSEL into an 87Rb CPT clock setup and demonstrate a clear CPT resonance, achieving ~2.5% contrast and ~3 kHz CPT linewidth (FWHM), corresponding to a measured figure of merit of ~8.2 ppm. We further map key operating tradeoffs relevant to clock optimization, including optical power broadening (contrast peaking near ~50–60 µW while CPT linewidth increases with optical power) and RF-drive dependence across representative modulation powers. These results connect device-level laser characteristics to clock-level observables and provide a manufacturable component pathway for scalable Rb-based timing modules.

Optical and Quantum Sensing and Precision Metrology II•Mar 5, 2026•0 citations

Application of vertical cavity surface emitting lasers in self-oscillating atomic clocks

Dmitry Strekalov, Andrey B. Matsko, Nan Yu +2 more

We discuss practical aspects of building atomic clocks consisting of an optoelectronic oscillator stabilized with the electromagnetically induced transparency resonance in a 87Rb atomic vapour cell, where a vertical cavity surface emitting laser (VCSEL) is used as a light source. In particular, we show how the variable microwave power injection into the VCSEL, its optical injection-locking, and the phase stability of the optical-microwave link influences the stability of the clocks.

Journal of Modern Optics•Oct 25, 2006•16 citations

Optical characterization of a vertical cavity surface emitting laser for a coherent population trapping frequency reference

Christopher M. Long, Kent D. Choquette

Characterization of a vertical cavity surface emitting laser source for a frequency reference that uses coherent population trapping with atomic rubidium is reported. The frequency reference requires a low-noise laser that operates in a single transverse mode at 794.7nm. We show that biasing the laser at low output power and modulating it at higher frequency than its low relaxation oscillation frequency are necessary to obtain symmetric modulation sidebands and improve the long-term performance of the frequency reference.

Journal of Applied Physics•Feb 1, 2008•18 citations
A new generation of small low-power atomic sensors, including clocks, magnetometers, and gyroscopes, is being developed based on recently available MEMS and VCSEL technologies.These sensors rely on spectroscopic interrogation of alkali atoms, typically rubidium or cesium, contained in small vapor cells.The relevant spectroscopic wavelengths (in vacuum) are 894.6 nm (D1) and 852.3 nm (D2) for cesium, and 795.0 nm (D1) and 780.2 nm (D2) for rubidium.The D1 wavelengths are either preferred or required, depending on the application, and vertical-cavity surface-emitting lasers (VCSELs) are preferred optical sources because of their low power consumption and circular output beam.This paper describes the required VCSEL characteristics for atomic clocks and magnetometers.The fundamental VCSEL requirement is single-frequency output with tunability to the particular spectroscopic line of interest.Singlepolarization and single-transverse-mode operation are implicit requirements.VCSEL amplitude noise and frequency noise are also important because they contribute significantly to the sensor signal-to-noise ratio.Additional desired VCSEL attributes are low cost, low power consumption, and several years of continuous operating lifetime.This paper also describes the 894-nm VCSELs that we have developed for cesium-based atomic sensors.In particular, we discuss VCSEL noise measurements and accelerated lifetime testing.Finally, we report the performance of prototype atomic clocks employing VCSELs.
The spectroscopic technique of coherent population trapping (CPT) enables an all-optical interrogation of the groundstate hyperfine splitting of cesium (or rubidium), compared to the optical-microwave double resonance technique conventionally employed in atomic frequency standards. All-optical interrogation enables the reduction of the size and power consumption of an atomic clock by two orders of magnitude, and vertical-cavity surface-emitting lasers (VCSELs) are preferred optical sources due to their low power consumption and circular output beam. Several research teams are currently using VCSELs for DARPA's chip-scale atomic clock (CSAC) program with the goal of producing an atomic clock having a volume < 1 cm^3, a power consumption < 30 mW, and an instability (Allan deviation) < 1x10^-11 during a 1-hour averaging interval. This paper describes the VCSEL requirements for CPT-based atomic clocks, which include single mode operation, single polarization operation, modulation bandwidth > 4 GHz, low power consumption (for the CSAC), narrow linewidth, and low relative intensity noise (RIN). A significant manufacturing challenge is to reproducibly obtain the required wavelength at the specified VCSEL operating temperature and drive current. Data are presented that show the advantage of operating at the D1 (rather than D2) resonance of the alkali atoms. Measurements of VCSEL linewidth will be discussed in particular, since atomic clock performance is especially sensitive to this parameter.
Laser phase noise (PM) to transmitted intensity noise (AM) conversion is a fundamental aspect of the field/matter interaction, and a major noise process that limits the frequency stability of next-generation atomic clocks. It arises from the absorption cross section’s dependence on laser phase and is especially pernicious for vertical-cavity surface-emitting lasers (VCSELs) given their relatively broad singlemode linewidths (e.g., 40 MHz). This is unfortunate, since VCSELs (with their very low operating power and low probability for mode hops) are otherwise ideal for quantum devices. Here, we demonstrate that atomic spectroscopy with N VCSELs reduces the total PM-to-AM noise by a factor of ~N1/2. Consequently, atomic clocks with (for example) five VCSELs would have a signal-to-noise ratio (SNR5) approximately 120% higher than the same device operating with one VCSEL (SNR1) at the same total power: SNR5/SNR1≌ 2.2.
We have undertaken the development of a chip-scale atomic clock (CSAC) whose design goals include short-term stability, sigma(sub y)(tau=1 hour), of 1x10(exp 11) with a total power consumption of 30 mW and an overall device volume of 1 cu cm. The stringent power requirement dominates the physics package architecture, dictating a small (<5 mm(exp 3) gaseous atomic ensemble interrogated by a low-power semiconductor laser. At PTTI 2002, we reported on initial experimental investigations leading to the decision to employ the coherent population trapping (CPT) interrogation technique. This paper describes our further progress on the CSAC effort, including the development of custom vertical cavity surface emitting laser (VCSEL) sources and techniques for microfabricating miniature cesium vapor cells comprised of anodically bonded silicon and glass. Measurements of the signal contrast and linewidth of both the cesium D1 and D2 resonance transitions are compared, and frequency stability measurements of the CSAC testbed are presented.
We report system-level evaluation of manufacturable, quantum-optimized 795-nm vertical-cavity surface-emitting lasers (Q-VCSELs) for rubidium (Rb) atomic sensors, with emphasis on integration into an 87Rb coherent population trapping (CPT) clock prototype. Operation at 795 nm (Rb D1) is constrained by increased internal loss/absorption, polarization instability, and stringent requirements on spectral purity. At the component level, we employ epitaxial and layout engineering to improve mode discrimination, polarization robustness, and high-temperature efficiency, achieving stable single-transverse-mode emission and linear polarization from 20–80 °C with side-mode suppression ratio (SMSR) >30 dB and polarization extinction ratio (PER) >20 dB. The devices exhibit low threshold current (<400 µA at 80 °C) and slope efficiency >0.5 mW/mA, enabling >1 mW single-mode, single-polarization output power in chip-scale form factors. Quantum-relevant characterization shows linewidth on the order of ~30–35 MHz and relative intensity noise below −130 dB/Hz at frequencies greater than 1KHz. We then integrate the Q-VCSEL into an 87Rb CPT clock setup and demonstrate a clear CPT resonance, achieving ~2.5% contrast and ~3 kHz CPT linewidth (FWHM), corresponding to a measured figure of merit of ~8.2 ppm. We further map key operating tradeoffs relevant to clock optimization, including optical power broadening (contrast peaking near ~50–60 µW while CPT linewidth increases with optical power) and RF-drive dependence across representative modulation powers. These results connect device-level laser characteristics to clock-level observables and provide a manufacturable component pathway for scalable Rb-based timing modules.
Characterization of a vertical cavity surface emitting laser source for a frequency reference that uses coherent population trapping with atomic rubidium is reported. The frequency reference requires a low-noise laser that operates in a single transverse mode at 794.7nm. We show that biasing the laser at low output power and modulating it at higher frequency than its low relaxation oscillation frequency are necessary to obtain symmetric modulation sidebands and improve the long-term performance of the frequency reference.