Steps Toward a Prototype Atomic Clock

1. Steps Toward a Prototype Atomic Clock Nathan Belcher Senior Research Midterm Talk 12.13.07

2. Acknowledgements • Prof. Irina Novikova • Prof. Eugeniy Mikhailov • Chris Carlin

3. Outline • Overall goal of project • Background • Hardware • DAVLL • Data and results • Future work

4. Overall Goal of Project • To create a prototype atomic clock • Need: • Laser • Transition • Laser Frequency Locking System • Counter

5. Overall Goal of Project continued • Have: • Vertical-cavity surface-emitting laser (VCSEL) • Transition in rubidium-87 --> 5S_1/2 to 5P_1/2 (794.7 nm) • Dichroic atomic vapor laser locking (DAVLL) • Counter

6. Outline • Overall goal of project • Background • Hardware • DAVLL • Data and results • Future work

7. Background • Atomic transition in rubidium-87 • 5S_1/2 to 5P_1/2

8. Background continued • Use electromagnetically induced transparency to measure transmitted light • The closer to hyperfine splitting resonance, the more transmission • Counter locked to maximum transmission which corresponds to clock frequency

9. Background continued • Lambda system requires two fields at different frequencies • Problem: inherent in lasers are small random shifts in frequency around a set frequency (“jumps”) • Bigger problem: if two lasers are physically separate, the “jumps” are random

10. Background continued • Solution: use phase modulation to create two fields out of one physical laser • Why? Both fields “jump” with each other so relative frequency can be set by external generator • Creates carrier with sideband comb

11. Background continued • Three phase modulation formulas

12. Background continued • Carrier to sideband spacing determined by input frequency

13. Outline • Overall goal of project • Background • Hardware • DAVLL • Data and results • Future work

14. Hardware • VCSEL

15. Hardware continued • Temperature stability

16. Hardware continued • VCSEL constant current source

17. Hardware continued • Measuring setups • Phase modulation measurement

18. Hardware continued • Fabry-Perot cavity • Second mirror moved by piezoelectric • Piezoelectric controlled by voltage source, with low frequency modulation

19. Hardware continued • Another measuring setup • Finding rubidium resonances • Vapor cell can be either isotropically pure Rb-87 or natural abundance Rb-85 and -87

20. Hardware continued • Modulators • Commercial digital synthesizer (Agilent E8275D) • Up to 14 dBm of power at very precise frequencies • Stellex Mini-YIG crystal oscillator • Current controlled tunable crystal with frequencies ranging from 5.95 GHz to 7.15 GHz at 15 dBm

21. Hardware continued • Inside of Stellex oscillator

22. Hardware continued • Stellex oscillator calibration measurements

23. Hardware continued • Solenoid and shields • External non-homogeneous fields that interact with vapor cell, shifting state frequencies • Need to control field vapor cell feels, so surround cell with solenoid to produce constant homogeneous magnetic field and shields to limit outside magnetic fields

24. Hardware continued

25. Outline • Overall goal of project • Background • Hardware • DAVLL • Data and results • Future work

26. DAVLL • Allows locking of frequency of VCSEL to specific rubidium resonance frequency

27. DAVLL continued • Optical hardware

28. DAVLL continued • Absorption and differential spectra

29. DAVLL continued • Electronics are used to amplify the raw signal from the optics and adjust the current to the laser accordingly • Example: if laser’s frequency is higher than zero point on raw signal, electronics will supply less current so frequency decreases

30. DAVLL continued • Flow chart of electronics

31. DAVLL continued • There have been issues with the raw signal • Raw signal offset with respect to zero point • This offset causes tension between raw signal and feedback to laser • Makes us question whether laser locked on resonance or if signals not representative of locking

32. DAVLL continued • Auxiliary input useful to find correct range laser needs to be in for resonances • Some work done to make voltage scales of function generator and raw signal approximately equivalent • This will allow us to compare Aux and Raw Signals to make sure locked on resonance

33. Outline • Overall goal of project • Background • Hardware • DAVLL • Data and results • Future work

34. Data and Results • VCSEL Modulation • Used Agilent E8275D and Stellex oscillator • With Agilent, saw 1-to-1 sideband-to-carrier ratio at 14 dBm at 6.834 GHz • With Stellex, saw sidebands greater than carrier at 15 dBm at 6.834 GHz

35. Data and Results continued • Agilent modulation

36. Data and Results continued • Stellex modulation

37. Data and Results continued • Rubidium resonances • Two vapor cells: isotropically pure Rb-87 and natural abundance Rb-85 and -87 • Pure Rb-87 cell inside shields, used to achieve EIT • Mixed Rb-85 and -87 cell used in DAVLL as reference cell

38. Data and Results continued • Isotropically pure Rb-87 cell

39. Data and Results continued • Natural abundance Rb-85 and -87 cell

40. Data and Results continued • Electromagnetically induced transparency (EIT) • Occurs when all electrons are driven to ‘dark’ state that does not interact with either electromagnetic field • Laser has 100% transmission

41. Data and Results continued • EIT in Irina’s pure vapor cell

42. Data and Results continued • EIT in my pure vapor cell

43. Future Work • True DAVLL locking • Understand offset issues and work through them • Add counter to system • Test clock against frequency from Agilent