MEMS Atomic Clocks for Timing, Frequency Control & Communications

Seventh Framework Programme
Founding schema: Small and medium-scale focused research projects
Call : ICT-2-3.6/Micro and Nanosystems

1 September 2008 - 29 février 2012
Grant agreement no.: 224132

Coordinator : Dr Christophe Gorecki
FEMTO-ST - Université de Franche-Comté (UFC-P5) - +33 381666607

Administrative Contact : Pauline Fournier
EU Affairs Manager
SAIC - Université de Franche-Comté - + 33 381665814


Atomic clocks provide enhanced accuracy, stability, and timing precision compared to quartz-based technologies. However, the size and power consumption of existing atomic clocks far exceeds those of quartz-based clocks, preventing their deployment in portable applications.
The technology of Micro Electro Mechanical Systems (MEMS), with its ability to shrink mechanical features and mechanisms down to micron scales, already provides substantial size and power reduction for applications spanning wireless communications, sensors, and fluidic systems, and is now emerging to provide similar advantages for frequency and timing references.
To this aim, MAC-TFC brings together a consortium made of five major academic institutions, two research institutes and three industrial partners.

The goal of the MAC-TFC proposal is to develop and demonstrate all the necessary technology to achieve an ultra-miniaturised, low-power caesium atomic clock, operating on the power of an AA battery, with less than 200 mW power consumption. The main application of miniature clock will be the wireless  synchronisation.

Technical challenges

To achieve the objective, the main technological challenges are :

- the technology of MEMS cell and the alkali-atom vapour filling procedurecritical to obtain a pure and stable atmosphere of Cs vapour with well controlled pressure of buffer gases
- the study of scaling limits in realisation of MEMS atomic clocks
- the respect of the low fixed power budget and portability
- the customisation of the VCSELs for CPT operation of the MEMS atomic clocks
- the realisation of the control electronics, interrogating the atomic resonator
MEMS Atomic Clocks for Timing, Frequency control and Communications


Today, atomic clocks are stable frequency sources and are important in civil and military applications ranging from communication systems to global positioning as well as synchronization of communication networks. During the last few years, considerable work has been carried out  by different US groups (NIST, Symmetricom) awarded by DARPA Chip Scale Atomic Clock program. The principle of proposed miniature clocks is based on Coherent Population Trapping (CPT), achieved in an extremely compact sealed vacuum cell of few cubic millimeters. These cells contain alkali vapors and are illuminated by a high-frequency modulated laser beam. The frequency stability of such atomic clocks is based on transitions between the well-defined ground state hyperfine levels of alkali atoms such as cesium or rubidium (Fig. 1).  In this solution, as opposed to classical rubidium vapour cell clocks, the CPT frequency standard does not require a microwave cavity to probe the atomic resonance. This permits the very compact physical package of the micro-clock, as shown in Fig. 2

MEMS Atomic Clocks for Timing, Frequency control and Communications
MEMS Atomic Clocks for Timing, Frequency control and Communications
Fig. 1
Fig. 2

Several methods of filling the micromachined cell with alkali vapors and introducing the buffer gas have been reported, but pratically all these methods are based on the activation of Cs vapour prior to the sealing of microcell. We propose a filling method in which the conflict of anodic bonding process with cesium chemistry is fully avoided which makes the fabrication of atomic micro-clock simpler and improves its potential mass production. This patented solution uses a Cs dispenser commercially available from SAES Getters, making possible to heat locally the Cs dispenser inside the sealed micro-cell. It is made below temperature range causing degradation of micro-cavity as well as reducing the negative effects of Cs chemistry. This will improve the control of the internal cell atmosphere purity as well as the buffer gases stability.

* Li-A. Liew, S. Knappe, J. Moreland, H. Robinson, L.Hollberg and J. Kitching, “Microfabricated alkali vapor cells”, Appl. Phys. Lett. 84, 2694-2696, (2004).
* R. Lutwak, J. Deng, W. Riley, M. Varghese, J. Leblanc, G. Tepolt, M. Mescher, D.K. Serkland, K.M. Geib, and G.M. Peake, “The Chip-Scale Atomic Clock – Low-Power Physics Package”, in Proc. of PTTI,  December  2004, Washington DC,  pp. 339-354.
* L. Nieradko, C. Gorecki, A. Douahi, V. Giordano, J.C. Beugnot, J.Dziuban, M. Moraja, “ New approach of fabrication and dispensing of micromachined cesium vapor cell “ J. Micro/Nanolith. MEMS MOEMS 7, 033013, (August 6, 2008).

Scientific and technical objectives

The goal of the MAC-TFC proposal is to develop and demonstrate all the necessary technology to achieve an ultra-miniaturised, low-power, atomic time and frequency reference units featuring :
- A size reduction to 10 cm3
- A power consumption reduction to < 155 mW
- A frequency stability of 5x10-11 over 1 hour

Research issues include, among others :
- Temperature stability, magnetic shielding and hermetic sealing of alkali vapour
- Dispensing and activation of Cs vapours as well as gettering of the internal atmosphere of micro-cell
- Integration with customised vertical cavity surface emitting laser (VCSEL) and photo detector
- Assembling and 3D LTCC packaging
- Miniaturized low-power ASIC integration and phase locking electronics

The work is concentrated on the delivery of two alternative atomic resonator configurations :
- Transmissive version of atomic resonator of relatively big size based on a sandwich silicon membrane-glass cavity-silicon membrane
- Horizontal reflective version of atomic resonator based on a horizontal silicon cavity with adjustable optical path and one cover glass

The industrial partners (Swatch Group R&D, Oscilloquartz) will identify operational requirements for potential applications and their industrialization.

First achievements of the consortium :

During the year 1 of MAC-TFC proposal, the Consortium focused on establishing theoretical limits of MEMS atomic clocks and demonstrating practical design and fabrication feasibilities, optimizing the performances of the atomic resonator. In parallel, the developments of building blocks for the miniaturized clock started : a fully customised semiconductor laser, several innovative approaches of Cs cells, low-power ASIC for the analog RF electronics, and the preliminary version of LTCC-packaging as shown in the following gallery of images.

Gallery of images representative for the period :

Transmissive T-cell after activation of alkali vapor (UFC-P5)

Reflective cavity of R-cell (UFC-P5)

Transmissive version of glass I-cell (PWR)

Prototype of Cs min-pills (SAES)

View of experimental clock (UniNE)

Mnufactured LTCC platform pads for the VCSEL (VTT)

Polarization-stable VCSEL with a monolithic surface grating (UULM)

Achievements of the consortium for the second year :

During the year 2 of MAC-TFC proposal, the Consortium focused on the developments and finalization of building blocks for the miniaturized clock : a fully customised VCSEL, several technologies of Cs cells, low-power ASIC for the analog RF electronics and the characterisation of these components started. The LTCC-packaging is now ready for the assembling of components of MAC-TFC clock.

Gallery of images representative for the period :

Observation of a quadratic temperature dependence of the Cs ground state hyperfine resonance frequency (UNINE and UFC-P5).

Getter-integrated T-cells containing Ne 100 torr (UFC-P5 & SAES).

Flip-chip bondable VCSEL . Dimensions are 300 µm x 300 µm (UULM).

ASIC (2 x 2 mm) for high resolution 4.6 GHz frequency synthesis (EPFL & Swatch Group R&D).
The paper from A. Al-Samaneh, S. Renz, A. Strodl, W. Schwarz, D. Wahl, R. Michalzik : 'Polarization-stable single-mode VCSELs for Cs-based MEMS atomic clock applications' , has received the "Best Student Paper Award" in the conference of "Semiconductor Lasers and Laser Dynamics IV" in Photonics Europe 2010, Brussels, Belgium.

RF test board with bonded ASIC (EPFL).

Thermally controllable multilayer LTCC platforms for Cs cells (VTT).

Objectives for the third year :

The third year of MAC-TFC proposal is concentrated on the LTCC-packaging of the physic elements using the transmission version of the atomic resonator. The final assembly of the physics package together with the developed electronics are tested before the setting of basis for technology transfer and pre-industrialisation for potential applications such as wireless network synchronisation.

Gallery of images representative for the period :

UFC-P5 & SAES transmissive cell with getter integration on glass cover improving the quality of internal atmosphere.

VTT : LTCC Cs cell subsystem.

UFC –P5 : Alternative Physic Package  demonstrator .

Final achievements :

During the final step of MAC-TFC the fabrication and testing of individual building blocks of MEMS clock were completed and the assembling of a macroscopic and alternative physics package demonstrator has been made, including all the building blocks developed by the project. The alternative demonstrator was useful to rapidly validate the feasibility of the MAC-TFC concept while all required functionalities have been properly implemented. This alternative demonstrator was followed by the chip-level integration and the miniature LTCC-based package of atomic resonator. Performance limiting factors have been identified and ways for improvements have been proposed, opening the implementation of MAC-TFC industrial applications in wireless synchronisation.

Realization of MAC-TFC clock demonstrator :

a) macroscopic and dismountable physics package

b) miniature physics  package

Physics package assembled on vacuum package pins

Miniature prototype Physics Package assembly