Strontium optical lattice clocks
Jérôme Lodewyck
J. Lodewyck — Sr OLCs – RTG April 2015
1/34
Clock ensemble at SYRTE
2 strontium lattice clocks
J. Lodewyck, R. Le Targat,
J.-L. Robyr, S. Bilicki
1 mercury lattice clock
S. Bize, L. De Sarlo,
R. Tyumenev, M. Favier
Frequency combs (Ti-Sa, fiber)
Y. Lecoq, R. Le Targat, D. Nicolodi
3 atomic fountains
(1 Cs, 1 Cs mobile, 1 Cs/Rb)
J. Guéna, P. Rosenbusch, M.
Abgrall, D. Rovera, S. Bize,
P. Laurent
J. Lodewyck — Sr OLCs – RTG April 2015
2/34
Clock ensemble at SYRTE
2 strontium lattice clocks
J. Lodewyck, R. Le Targat,
J.-L. Robyr, S. Bilicki
1 mercury lattice clock
S. Bize, L. De Sarlo,
R. Tyumenev, M. Favier
Frequency combs (Ti-Sa, fiber)
Y. Lecoq, R. Le Targat, D. Nicolodi
3 atomic fountains
(1 Cs, 1 Cs mobile, 1 Cs/Rb)
J. Guéna, P. Rosenbusch, M.
Abgrall, D. Rovera, S. Bize,
P. Laurent
J. Lodewyck — Sr OLCs – RTG April 2015
2/34
Content
1
Atomic clocks
2
Strontium lattice clocks
3
Clock comparisons
4
Towards operational optical clocks
5
Prospects
J. Lodewyck — Sr OLCs – RTG April 2015
3/34
Content
1
Atomic clocks
2
Strontium lattice clocks
3
Clock comparisons
4
Towards operational optical clocks
5
Prospects
J. Lodewyck — Sr OLCs – RTG April 2015
4/34
History of atomic clocks accuracy
10-10
EssenandParry
H
10-11
Opticalfrequencycombs
Ca
Clocksaccuracy
10-12
H
10-13
H
RedefinitionoftheSIsecond
H
Hg+,Yb +,Ca
10-14
+
Sr ,Yb +
10-15
Sr
Hg+,Yb +
Atomicfountains
10-16
Sr
Al+
10-17
10-18
1950
Microwaveclocks
Opticalclocks
1960
1970
Hg+
Al+ Sr
Sr
1980
1990
2000
2010
2020
Microwave clocks: 1 order of magnitude every 10 years since 1950
Optical clocks: 2 orders of magnitude every 10 years since 1990
J. Lodewyck — Sr OLCs – RTG April 2015
5/34
Reminder about atomic clocks and notations
Atomic clock:
1: Local oscillator
2: Atomic resonance
|ei
ω(t)
J. Lodewyck — Sr OLCs – RTG April 2015
ω0
|f i
6/34
Reminder about atomic clocks and notations
Atomic clock:
1: Local oscillator
ω(t)
2: Atomic resonance
Lock of the LO on the atom:
ω(t) = ω0 (1 + + y (t))
= systematic frequency
shift
accuracy: uncertainty on J. Lodewyck — Sr OLCs – RTG April 2015
|ei
ω0
|f i
y (t) = frequency noise
stability σy (τ ): Allan deviation of
y (t)
6/34
Reminder about atomic clocks and notations
Atomic clock:
1: Local oscillator
ω(t)
2: Atomic resonance
Lock of the LO on the atom:
ω(t) = ω0 (1 + + y (t))
= systematic frequency
shift
accuracy: uncertainty on |ei
ω0
|f i
y (t) = frequency noise
stability σy (τ ): Allan deviation of
y (t)
Optical vs. microwave
Microwave clocks:
ω0 /2π ' 1010 Hz
Optical clocks:
ω0 /2π ' 1014 to 1015 Hz
J. Lodewyck — Sr OLCs – RTG April 2015
6/34
Reminder about atomic clocks and notations
Atomic clock:
1: Local oscillator
ω(t)
2: Atomic resonance
|ei
Lock of the LO on the atom:
ω(t) = ω0 (1 + + y (t))
= systematic frequency
shift
accuracy: uncertainty on Optical vs. microwave
Microwave clocks:
ω0 /2π ' 1010 Hz
Optical clocks:
ω0 /2π ' 1014 to 1015 Hz
J. Lodewyck — Sr OLCs – RTG April 2015
ω0
|f i
y (t) = frequency noise
stability σy (τ ): Allan deviation of
y (t)
Example: frequency perturbation δω/2π = 1 mHz
Microwave clocks:
Optical clocks:
δω
ω0
δω
ω0
= 10−13
= 10−18
⇒ improves both the accuracy and the frequency
stability
6/34
Going to optical:
improving the frequency stability
Quantum Projection Noise: ultimate frequency stability:
r
1
1
Tc
σy (τ ) ' √
Q Natom
τ
where
Nat is the number of probed atoms
Tc is the cycle time
Q is the quality factor of the resonance
ω0
Q=
∝ Ti ω0
∆ω
J. Lodewyck — Sr OLCs – RTG April 2015
7/34
Going to optical:
improving the frequency stability
Quantum Projection Noise: ultimate frequency stability:
r
1
1
Tc
σy (τ ) ' √
Q Natom
τ
where
Nat is the number of probed atoms
Tc is the cycle time
Q is the quality factor of the resonance
ω0
Q=
∝ Ti ω0
∆ω
How optical clocks help?
Large clock frequency ω0
(optical transition)
Large interaction time Ti
(trapped atoms)
But Nat usually smaller
J. Lodewyck — Sr OLCs – RTG April 2015
7/34
Going to optical:
improving the frequency stability
Quantum Projection Noise: ultimate frequency stability:
r
1
1
Tc
σy (τ ) ' √
Q Natom
τ
where
Nat is the number of probed atoms
Tc is the cycle time
Q is the quality factor of the resonance
ω0
Q=
∝ Ti ω0
∆ω
How optical clocks help?
Examples of QPN stabilities
Large clock frequency ω0
(optical transition)
Cesium fountain:
σy (1 s) = 2 × 10−14
Large interaction time Ti
(trapped atoms)
Ion clock: σy (1 s) = 10−15
But Nat usually smaller
J. Lodewyck — Sr OLCs – RTG April 2015
Lattice clock: σy (1 s) = 10−17
(not demonstrated yet)
7/34
Going to optical: improving the accuracy
Accuracy budget of the FO2 cesium microwave fountain:
Effect
Quadratic Zeeman shift
Collisions and cavity pulling
Spectral purity (spurious sidebands),
leakage, synchronous phase modulation
Raby and Ramsey pulling
Background collisions
Black-body radiation shift
Distributed cavity phase shift
(Residual 1st order Doppler)
Microwave lensing (Microwave recoil)
Second order Doppler shift
Total
Correction (in 10−16 )
-1915.9
112.0
0
Uncertainty (in 10−16 )
0.3
1.2
< 0.5
0
0
168.0
-0.9
<0.1
1.0
0.6
0.9
-0.7
0
-1637.5
0.7
0.1
2.1
2 categories of systematic effects:
Effects (mostly) independent of the frequency
Effects proportional to the frequency
J. Lodewyck — Sr OLCs – RTG April 2015
Guéna et al., Progress in Atomic
Fountains at LNE-SYRTE,IEEE
UFFC, 59, 391-420 (2012)
8/34
Content
1
Atomic clocks
2
Strontium lattice clocks
3
Clock comparisons
4
Towards operational optical clocks
5
Prospects
J. Lodewyck — Sr OLCs – RTG April 2015
9/34
Optical lattice clocks
Atoms loaded from a MOT to an optical lattice
formed by a 1D standing wave
Probing a narrow optical resonance with an
ultra-stable “clock” laser
Stabilize the clock laser on the narrow
resonance
J. Lodewyck — Sr OLCs – RTG April 2015
10/34
Optical lattice clocks
Atoms loaded from a MOT to an optical lattice
formed by a 1D standing wave
Probing a narrow optical resonance with an
ultra-stable “clock” laser
Stabilize the clock laser on the narrow
resonance
Combine several advantages:
Optical clock
Lamb-Dicke regime
insensitive to motional effects
⇒
Record frequency stability
Record accuracy
Large number of atoms
J. Lodewyck — Sr OLCs – RTG April 2015
10/34
Magic wavelength
Light-shift
Intense trapping laser
⇒ light-shift of several MHz
At the magic wavelength, this
huge light-shift is canceled
J. Lodewyck — Sr OLCs – RTG April 2015
11/34
Magic wavelength
Light-shift
Intense trapping laser
⇒ light-shift of several MHz
At the magic wavelength, this
huge light-shift is canceled
Looking closer. . .
Polarization dependent 1st order light-shifts (vector and tensor shifts)
Mostly canceled for a J = 0 → J = 0 transition.
Hyperpolarizability (2 photon transitions)
Multipolar effects (E2/M1)
⇒ Need to pay attention to achieve high accuracy
P. Westergaard et al., PRL 106 210801 (2011)
J. Lodewyck — Sr OLCs – RTG April 2015
11/34
Magic wavelength
2 ,0
1 ,4
R e s id u a l p o la r is a b ility : 3 .9 ( 5 ) m H z /E
Residual polarisability: -0.41 (27) mHz/Er
2
Hyperpolarisability: 0.44 (10) µHz/Er
r
1 ,0
Sr 2:
0 ,8
0 ,6
0 ,4
0 ,2
L a ttic e in d u c e d lig h t s h ift ( H z )
Sr 1:
L a ttic e in d u c e d s h ift ( H z )
1 ,2
1 ,5
1 ,0
0 ,5
0 ,0
0 ,0
0
5 0
1 0 0
1 5 0
L a ttic e D e p th [E r]
2 0 0
2 5 0
3 0 0
0
5 0 0
1 0 0 0
1 5 0 0
Lattice depth [Er]
2 0 0 0
2 5 0 0
Looking closer. . .
Polarization dependent 1st order light-shifts (vector and tensor shifts)
Mostly canceled for a J = 0 → J = 0 transition.
Hyperpolarizability (2 photon transitions)
Multipolar effects (E2/M1)
⇒ Need to pay attention to achieve high accuracy
P. Westergaard et al., PRL 106 210801 (2011)
J. Lodewyck — Sr OLCs – RTG April 2015
11/34
Why strontium ?
³S₁
Sr lattice clock
Wavelengths accessible with
semi-conductor sources
Magic wavelength at 813 nm
Implemented in many laboratories
⇒ good candidate for a new SI second
¹P₁
λ = 688 nm
MOT
³P₀
Clock
λ = 461 nm
Γ = 32 MHz
³P₁
³P₂
λ = 689 nm
Γ = 7.6 kHz
¹S₀
J. Lodewyck — Sr OLCs – RTG April 2015
12/34
Why strontium ?
³S₁
Sr lattice clock
Wavelengths accessible with
semi-conductor sources
¹P₁
Magic wavelength at 813 nm
Implemented in many laboratories
⇒ good candidate for a new SI second
λ = 688 nm
MOT
³P₀
Clock
λ = 461 nm
Γ = 32 MHz
³P₁
³P₂
λ = 689 nm
Γ = 7.6 kHz
¹S₀
Other candidates
Yb: mostly equivalent to Sr
Hg: lower systematic effects, but UV lasers
Mg: here !
J. Lodewyck — Sr OLCs – RTG April 2015
12/34
Clock sequence
1
“Blue MOT”: a few 106
atoms of 87 Sr at 3 mK
Optical lattice of depth
100 µK
³S₁
¹P₁
Blue MOT
461 nm
³P₀
³P₁
³P₂
¹S₀
J. Lodewyck — Sr OLCs – RTG April 2015
13/34
Clock sequence
1
“Blue MOT”: a few 106
atoms of 87 Sr at 3 mK
Optical lattice of depth
100 µK
“Drain” the atoms to
metastable states
⇒ 104 atoms trapped in
300 ms
³S₁
¹P₁
Drain 2
688 nm
Blue MOT
461 nm
³P₀
³P₁
³P₂
Drain 1
689 nm
¹S₀
J. Lodewyck — Sr OLCs – RTG April 2015
13/34
Clock sequence
1
2
“Blue MOT”: a few 106
atoms of 87 Sr at 3 mK
Optical lattice of depth
100 µK
“Drain” the atoms to
metastable states
⇒ 104 atoms trapped in
300 ms
³S₁
Repump 2
707 nm
¹P₁
Repump 1
679 nm
³P₀
Repumping
³P₁
³P₂
¹S₀
J. Lodewyck — Sr OLCs – RTG April 2015
13/34
Clock sequence
1
“Blue MOT”: a few 106
atoms of 87 Sr at 3 mK
Optical lattice of depth
100 µK
“Drain” the atoms to
metastable states
⇒ 104 atoms trapped in
300 ms
2
Repumping
3
Narrow line-width cooling
³S₁
¹P₁
³P₀
³P₁
³P₂
Cooling
689 nm
¹S₀
J. Lodewyck — Sr OLCs – RTG April 2015
13/34
Clock sequence
1
“Blue MOT”: a few 106
atoms of 87 Sr at 3 mK
Optical lattice of depth
100 µK
“Drain” the atoms to
metastable states
⇒ 104 atoms trapped in
300 ms
2
Repumping
3
Narrow line-width cooling
4
Optical pumping to mF = ±9/2
J. Lodewyck — Sr OLCs – RTG April 2015
³S₁
¹P₁
³P₀
³P₁
³P₂
Optical
pumping
689 nm
¹S₀
mF = ±9/2
13/34
Clock sequence
1
“Blue MOT”: a few 106
atoms of 87 Sr at 3 mK
Optical lattice of depth
100 µK
“Drain” the atoms to
metastable states
⇒ 104 atoms trapped in
300 ms
2
Repumping
3
Narrow line-width cooling
4
Optical pumping to mF = ±9/2
5
Clock interrogation
J. Lodewyck — Sr OLCs – RTG April 2015
³S₁
¹P₁
³P₀
³P₁
³P₂
Clock
698 nm
¹S₀
13/34
Clock sequence
1
“Blue MOT”: a few 106
atoms of 87 Sr at 3 mK
Optical lattice of depth
100 µK
“Drain” the atoms to
metastable states
⇒ 104 atoms trapped in
300 ms
2
Repumping
3
Narrow line-width cooling
4
Optical pumping to mF = ±9/2
5
Clock interrogation
6
Normalized detection by
fluorescence
J. Lodewyck — Sr OLCs – RTG April 2015
³S₁
¹P₁
Blue probe
461 nm
³P₀
³P₁
³P₂
¹S₀
13/34
Sr clock laser
10 cm horizontal ULE spacer and silica mirrors
Not at inversion of CTE ⇒ 3 layers of thermal shielding
(short term temperature fluctuations in the nK range, τ = 4 days)
Residual drift of a few 10s of mHz/s
J. Lodewyck — Sr OLCs – RTG April 2015
(feed forward compensation below 1mHz/s)
14/34
Locking the clock laser to the atomic
transition
0.9
Transition probability
0.8
0.7
0.6
0.5
3 Hz
0.4
0.3
0.2
0.1
0
−8
−6
−4
−2
0
2
4
6
8
Frequency (Hz)
Resonance of the clock transition
Fourier limited at 3 Hz (250 ms)
Laser noise dominating
J. Lodewyck — Sr OLCs – RTG April 2015
15/34
Fractional allan deviation
Locking the clock laser to the atomic
transition
0.9
Transition probability
0.8
0.7
0.6
0.5
3 Hz
0.4
0.3
10 -14
10 -15
Sr 1 - cavity
Sr 2 - cavity
Sr 2 - Sr 1
3.13e-15 at 1 s
10 -16
10 -17
0.1
1
10
100
1000
10000 100000
Time (s)
0.2
Atoms vs cavity
0.1
0
−8
−6
−4
−2
0
2
4
6
8
Frequency (Hz)
Resonance of the clock transition
Fourier limited at 3 Hz (250 ms)
Laser noise dominating
J. Lodewyck — Sr OLCs – RTG April 2015
< 5 s : limited by the atoms
√
(Dick effect) 2.2 × 10−15 / τ
> 5 s : limited by the thermal
noise of the cavity 8 × 10−16
Clock comparison
√
3.1 × 10−15 / τ
resolution in the low 10−17
15/34
Measuring systematic effects
Comparing two clocks.
Differential measurement on a single clock using the ultra-stable laser
as a flywheel.
Example: cold collision
10-14
0.01
0.006
10-15
Frequency[Hz]
FractionalAllandeviation
0.008
10-16
0.004
0.002
0
-0.002
10-17
-0.004
10-18
0.1
2.9e-15/√τ
1
-0.006
10
100
Time(s)
1000
10000
100000
3
3.2
3.4
3.6
3.8
4
High density (3 atoms per site) vs. Low density (< 1 atom per site)
J. Lodewyck — Sr OLCs – RTG April 2015
16/34
Accuracy budget for the Sr clocks
Main contributions: (Sr 2)
Effet
BBR
Quadratic Zeeman
Lattice light-shift
Density shift
Line Pulling
Probe ligth-shift
AOM phase chirp
Total
Correction
5.208 × 10−15
1.317 × 10−15
−3 × 10−17
0
0
4 × 10−19
−8 × 10−18
648.74 × 10−17
Uncertainty
6.7 × 10−17
1.3 × 10−17
2 × 10−17
1.7 × 10−17
2 × 10−17
4 × 10−19
8 × 10−18
7.6 × 10−17
The black-body shift remains the most important contribution
Lattice light-shift and cold collisions particularly relevant.
J. Lodewyck — Sr OLCs – RTG April 2015
17/34
Unexpected systematic shift: DC stark shift
First comparison between our two Sr clocks: 10−13 discrepancy !
→ due to the presence of static charges on the view-ports.
Sr2 - Sr1 (frac. units, x 10-14)
0
-8.6
-8.8
-9
-9.2
-9.4
-2
-4
0
5
10
15
20
-6
-8
τ = 264 days
-10
0
20
40
60
80
Time (days)
J. Lodewyck — Sr OLCs – RTG April 2015
100
120
18/34
Unexpected systematic shift: DC stark shift
First comparison between our two Sr clocks: 10−13 discrepancy !
→ due to the presence of static charges on the view-ports.
→ charges extracted by the photo-electric effect induced by UV light
0
-4
-1
-14
-2
)
-8.6
-8.8
-9
-9.2
-9.4
0
5
10
15
20
-6
-8
Sr2 - Sr1 (frac. units x 10
Sr2 - Sr1 (frac. units, x 10-14)
0
-2
UV
-3
UV
UV
UV
-4
-5
-6
τ = 264 days
-10
0
20
40
60
80
Time (days)
J. Lodewyck — Sr OLCs – RTG April 2015
100
120
-7
0
1000 2000 3000 4000 5000 6000 7000 8000 9000
Time (s)
18/34
Unexpected systematic shift: DC stark shift
First comparison between our two Sr clocks: 10−13 discrepancy !
→ due to the presence of static charges on the view-ports.
→ charges extracted by the photo-electric effect induced by UV light
Residual charges characterized with an electrode
±V
Electrode
−E
ν
E
−E~
∆ν
Atoms
E~
J. Lodewyck — Sr OLCs – RTG April 2015
18/34
Unexpected systematic shift: DC stark shift
First comparison between our two Sr clocks: 10−13 discrepancy !
→ due to the presence of static charges on the view-ports.
→ charges extracted by the photo-electric effect induced by UV light
Residual charges characterized with an electrode
±V
Electrode
ν
δE
−E
−E~
E
δν
~
δE
Atoms
∆ν+
∆ν−
E~
⇒ no residual charge at the level of 1.5 × 10−18
J. Lodewyck — Sr OLCs – RTG April 2015
18/34
Other groups
Rapid progress for optical lattice clocks
JILA: Sr optical lattice clock with 2 × 10−18 accuracy
NIST: Stability down to 1.6 × 10−18 after 7 h between 2 Yb clocks
PTB: Sr optical lattice clock with 3 × 10−17 accuracy
ultra-stable laser with 8 × 10−17 noise floor
Riken: Comparison between to cryogenic Sr clocks with 7.2 × 10−18
accuracy.
J. Lodewyck — Sr OLCs – RTG April 2015
19/34
Content
1
Atomic clocks
2
Strontium lattice clocks
3
Clock comparisons
4
Towards operational optical clocks
5
Prospects
J. Lodewyck — Sr OLCs – RTG April 2015
20/34
Clock comparisons: why and how ?
Objectives of clock comparisons
Prove the reproducibility of optical lattice clocks
Determine and track frequency ratios between different atomic species
Measure offsets and variations of the geo-potential at
country/continental scales
Long term goal: use optical clocks to define international time scales
J. Lodewyck — Sr OLCs – RTG April 2015
21/34
Clock comparisons: why and how ?
Objectives of clock comparisons
Prove the reproducibility of optical lattice clocks
Determine and track frequency ratios between different atomic species
Measure offsets and variations of the geo-potential at
country/continental scales
Long term goal: use optical clocks to define international time scales
Means of comparison
Local or local area comparisons
⇒ limited range
Satellite comparisons (GPS/TWSTFT)
⇒ limited resolution (fine for microwave clocks, but not optical)
Stabilized optical fiber links
⇒ limited to continental scales and connected places
Pharao/ACES space clock on board the ISS (2017)
⇒ limited in time
J. Lodewyck — Sr OLCs – RTG April 2015
21/34
Frequency difference Sr2-Sr1 (Hz)
Local Sr vs. Sr comparison
0,2
-16
4x10
0,1
-16
2x10
0,0
-0,1
0
Sr2-Sr1
Statistic
Average
Statistic
Systematics
Systematics
1
2
3
4
5
6
Measurement #
7
-16
-2x10
8
Set of comparisons with a small statistical error bar
Agreement at the 10−16 level:
Sr2 - Sr1 = 1.1 × 10−16 ± 2 × 10−17 (stat) ± 1.6 × 10−16 (sys)
Only after unexpected systematic effects have been discovered
(DC Stark shift, TA spontaneous emission)
R. Le Targat et al. Nat. Commun. 4 2109 (2013)
J. Lodewyck — Sr OLCs – RTG April 2015
22/34
Sr vs microwave atomic fountains: stability
Absolute frequency measurements
10-13
Sr vs Rb
(best µWave vs optical stability)
10-13
-14
10
FractionalAllandeviation
FractionalAllandeviation
Sr vs Cs and Rb
10-15
10-16
3.4e-14/√τ
3.9e-14/√τ
10-17
0.1
1
10
100
10-14
10-15
1000 10000100000 1x106
Time(s)
2.8e-14/√τ
10-16
1
10
100
1000
10000
Time(s)
Frequency stability limited by the QPN of the microwave fountains
10−16 resolution after 12 h
mid 10−17 resolution after 7 days
J. Lodewyck — Sr OLCs – RTG April 2015
23/34
Sr vs microwave atomic fountains: accuracy
Sept. 2014
3.6
Oct. 2014
3.4
Mar. 2015
5x10
-16
0
3.2
-5x10
-16
-1x10
-15
SYRTE 2011 (Nat. Comm. 2013)
3
2.8
-1.5x10
-15
2.6
-2x10
[f(Sr)/f(X)]/[f(Sr, SI)/f(X, SI)] - 1
f(Sr) - 429228004229870 Hz
SI recommendation
-15
FO1
2.4
FO2-Cs
FO2-Rb
-2.5x10
-15
Each measurement limited by the accuracy and stability of the
fountains
Good repeatability
J. Lodewyck — Sr OLCs – RTG April 2015
24/34
Variation of fundamental constants
10
f(Sr) - 429228004229870 Hz
8
SYRTE
JILA
Tokyo
PTB
NICT
NMIJ
6
4
10
2
15 x 10
-15
10 x 10
-15
5 x 10
-15
0 x 10
0
-15
-5 x 10
-15
f(Sr)/f(Sr, SI) - 1
Frequency ratio Sr vs. Cs
0
-10 x 10
-15
-2
53500
2005
54000
2006
54500
2007
2008
J. Lodewyck — Sr OLCs – RTG April 2015
55000
2009
55500
2010
56000
2011
2012
56500
2013
57000
2014
25/34
Variation of fundamental constants
10
f(Sr) - 429228004229870 Hz
8
SYRTE
JILA
Tokyo
PTB
NICT
NMIJ
6
4
10
2
15 x 10
-15
10 x 10
-15
5 x 10
-15
0 x 10
0
-15
-5 x 10
-15
f(Sr)/f(Sr, SI) - 1
Frequency ratio Sr vs. Cs
0
-10 x 10
-15
-2
53500
2005
54000
2006
54500
2007
2008
55000
2009
55500
2010
56000
2011
2012
56500
2013
57000
2014
Best agreed upon frequency measurement (SYRTE/PTB)
Bounds on drifts of fundamental constants
Anchor point for future measurements
(regular comparisons between Sr, Hg, Cs, Rb)
J. Lodewyck — Sr OLCs – RTG April 2015
25/34
Content
1
Atomic clocks
2
Strontium lattice clocks
3
Clock comparisons
4
Towards operational optical clocks
5
Prospects
J. Lodewyck — Sr OLCs – RTG April 2015
26/34
Improving the clock reliability
Motivation: Need for operational optical clocks
Enhancing the statistical resolution
⇒ better characterization of systematic effects
Participating to the Pharao/ACES space mission
Establishing time scales with optical clocks
J. Lodewyck — Sr OLCs – RTG April 2015
27/34
Improving the clock reliability
Motivation: Need for operational optical clocks
Enhancing the statistical resolution
⇒ better characterization of systematic effects
Participating to the Pharao/ACES space mission
Establishing time scales with optical clocks
Goal:
Reach a level of maturity equivalent to the Cs based clock architecture
⇒ possible redefinition of the SI second
J. Lodewyck — Sr OLCs – RTG April 2015
27/34
Long operation (Feb. 2014)
Attended operation of the full metrological chain for 100 consecutive hours
Strontium clocks
Frequency of cavity vs. strontium (dedrifted): 87 % uptime
J. Lodewyck — Sr OLCs – RTG April 2015
28/34
Long operation (Feb. 2014)
Attended operation of the full metrological chain for 100 consecutive hours
Strontium clocks
Frequency of cavity vs. strontium (dedrifted): 87 % uptime
Full chain
Frequency of strontium vs. maser: 80 % uptime
Identified breaking points
J. Lodewyck — Sr OLCs – RTG April 2015
28/34
Long operation (Oct. 2014)
SYRTE-Sr2
Day UTC Points Uptime
24/10 08:00 3387 94,08%
09:00 3375 93,75%
10:00 3455 95,97%
11:00 3416 94,89%
12:00 3460 96,11%
13:00 3505 97,36%
14:00 3501 97,25%
15:00 3476 96,56%
16:00 3595 99,86%
17:00 2032 56,44%
18:00 3595 99,86%
19:00 3597 99,92%
20:00 3598 99,94%
21:00 3596 99,89%
22:00 3597 99,92%
23:00 3599 99,97%
25/10 00:00 3599 99,97%
01:00 3600 100,00%
02:00 3600 100,00%
03:00 3600 100,00%
04:00 3593 99,81%
05:00 3588 99,67%
06:00 3599 99,97%
07:00 3601 100,03%
08:00 3594 99,83%
09:00 2732 75,89%
10:00 2565 71,25%
11:00 3599 99,97%
12:00 3592 99,78%
13:00 3599 99,97%
14:00 3591 99,75%
15:00 3600 100,00%
16:00 3600 100,00%
17:00 3599 99,97%
18:00 3599 99,97%
19:00 3601 100,03%
20:00 3080 85,56%
21:00 3600 100,00%
22:00 3596 99,89%
23:00 3599 99,97%
26/10 00:00 3090 85,83%
01:00
02:00
03:00
04:00
05:00
06:00
07:00
08:00
09:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
23:00
27/10 00:00
01:00
02:00
03:00
04:00
05:00
06:00
07:00
08:00
09:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
0
0
0
0
0
0
0
704
2766
3594
3580
3597
3594
2686
3199
3585
3264
3595
3593
3593
3596
3594
3596
3595
3598
3595
3594
3597
3590
3598
3374
3598
3587
3600
3597
3594
3569
3598
3590
3590
3600
3599
3592
0,00%
0,00%
0,00%
0,00%
0,00%
0,00%
0,00%
19,56%
76,83%
99,83%
99,44%
99,92%
99,83%
74,61%
88,86%
99,58%
90,67%
99,86%
99,81%
99,81%
99,89%
99,83%
99,89%
99,86%
99,94%
99,86%
99,83%
99,92%
99,72%
99,94%
93,72%
99,94%
99,64%
100,00%
99,92%
99,83%
99,14%
99,94%
99,72%
99,72%
100,00%
99,97%
99,78%
20:00
21:00
22:00
23:00
28/10 00:00
01:00
02:00
03:00
04:00
05:00
06:00
07:00
08:00
09:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
23:00
29/10 00:00
01:00
02:00
03:00
04:00
05:00
06:00
07:00
08:00
09:00
10:00
11:00
12:00
13:00
14:00
3596
3592
3480
3600
3597
3600
3599
3599
3600
3600
3599
3600
3559
3598
3571
3600
3599
3597
3599
3599
3598
3600
3598
3600
3597
3591
3599
3599
3569
3600
3600
3600
3600
3600
3600
3600
3594
3592
3600
3600
3600
3593
3599
99,89%
99,78%
96,67%
100,00%
99,92%
100,00%
99,97%
99,97%
100,00%
100,00%
99,97%
100,00%
98,86%
99,94%
99,19%
100,00%
99,97%
99,92%
99,97%
99,97%
99,94%
100,00%
99,94%
100,00%
99,92%
99,75%
99,97%
99,97%
99,14%
100,00%
100,00%
100,00%
100,00%
100,00%
100,00%
100,00%
99,83%
99,78%
100,00%
100,00%
100,00%
99,81%
99,97%
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
23:00
30/10 00:00
01:00
02:00
03:00
04:00
05:00
06:00
07:00
08:00
09:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
23:00
31/10 00:00
01:00
02:00
03:00
04:00
05:00
06:00
07:00
3596
3594
3600
3595
3594
3591
3422
3596
3600
3600
3600
3598
3599
3600
3594
3594
3600
3599
3600
2481
0
0
3298
3590
3574
3580
3597
3595
3596
3568
3589
3595
3598
3588
3594
3600
3597
3598
3592
3595
3593
99,89%
99,83%
100,00%
99,86%
99,83%
99,75%
95,06%
99,89%
100,00%
100,00%
100,00%
99,94%
99,97%
100,00%
99,83%
99,83%
100,00%
99,97%
100,00%
68,92%
0,00%
0,00%
91,61%
99,72%
99,28%
99,44%
99,92%
99,86%
99,89%
99,11%
99,69%
99,86%
99,94%
99,67%
99,83%
100,00%
99,92%
99,94%
99,78%
99,86%
99,81%
Total 558694 92,38%
FractionalAllandeviation
ITOC campaign: operation > 1 week
10-14
10-15
10-16
0.1
1
10
100
1000
10000 100000
Time(s)
Unattended at night, limited maintenance during the day
92% total uptime
7 consecutive hours without a single missing point
√
Improved stability at 10−15 / τ
J. Lodewyck — Sr OLCs – RTG April 2015
29/34
Content
1
Atomic clocks
2
Strontium lattice clocks
3
Clock comparisons
4
Towards operational optical clocks
5
Prospects
J. Lodewyck — Sr OLCs – RTG April 2015
30/34
Stability as an enabling factor
Stability enables a faster characterization of systematic effects
√
For OLC, the stability is currently limited to a few 10−16 / τ
(Dick effect: sampling of the clock laser noise)
J. Lodewyck — Sr OLCs – RTG April 2015
31/34
Stability as an enabling factor
Stability enables a faster characterization of systematic effects
√
For OLC, the stability is currently limited to a few 10−16 / τ
(Dick effect: sampling of the clock laser noise)
Prospects for lattice clock stability
√
QPN limit = a few 10−17 / τ for 104 atoms
√
QPN can be < 10−17 / τ with a large number of atoms
Quantum engineered states to overcome the QPN
J. Lodewyck — Sr OLCs – RTG April 2015
31/34
Stability as an enabling factor
Stability enables a faster characterization of systematic effects
√
For OLC, the stability is currently limited to a few 10−16 / τ
(Dick effect: sampling of the clock laser noise)
Prospects for lattice clock stability
√
QPN limit = a few 10−17 / τ for 104 atoms
√
QPN can be < 10−17 / τ with a large number of atoms
Quantum engineered states to overcome the QPN
Possible solutions
Better lasers (reduced thermal noise)
Long cavities
Cryogenic cavities
Crystalline coating
Spectral Hole Burning
Optimized sequence
Reduced dead time (Dick effect cancels for 0 dead-time)
Non destructive detection for faster preparation of the atoms
J. Lodewyck — Sr OLCs – RTG April 2015
31/34
Non destructive detection based on a
Mach-Zehnder
Aim : measure the transition
probability without loosing the atoms
by dispersive interaction
f
Local
Oscillator
f
EOM
Lattice
BPF f
Phase
signal
Probe
shot noise limited heterodyne
detection
SNR = 100 per shot
J. Lodewyck — Sr OLCs – RTG April 2015
32/34
Non destructive detection based on a
Mach-Zehnder
Aim : measure the transition
probability without loosing the atoms
by dispersive interaction
f
EOM
Probe
shot noise limited heterodyne
detection
Phase Noise (rad2/Hz)
Lattice
BPF f
Phase
signal
Local
Oscillator
Transition probability
f
0.6
0.5
0.4
0.3
0.2
0.1
RMS = 2%
0
−0.1
−300 −200 −100
10-10
10-11
-65
0
100
200
300
Detuning (kHz)
Noise at 125 Hz
Noise at 1 kHz
-60
-55 -50 -45 -40 -35
Detected power (dBm)
-30
-25
keeps 95% of atoms at 1000 ER
J. Lodewyck et al. PRA 79 061041(R) (2009)
SNR = 100 per shot
J. Lodewyck — Sr OLCs – RTG April 2015
32/34
Cavity based non destructive detection
813 nm
461 nm
Cavity based detection
Self-alignment
Increased SNR (∼
√
F)
Quantum regime for sub-QPN
stability
J. Lodewyck — Sr OLCs – RTG April 2015
33/34
Cavity based non destructive detection
813 nm
461 nm
Cavity based detection
Self-alignment
Increased SNR (∼
√
F)
Quantum regime for sub-QPN
stability
J. Lodewyck — Sr OLCs – RTG April 2015
Cavity finesse ' 20000
Heterodyne detection system
Reduced BBR uncertainty
33/34
Conclusion
Summary
Evaluation of the Sr accuracy budget
Absolute frequency measurements of Sr
Comparisons of two state-of-the-art Sr clocks
→ essential to fully confirm the accuracy budget
Ongoing international clock comparisons
Outlook
Control of the BBR for Sr by controlling the temperature of the
environment
Cavity based non destructive detection of the transition probability
(Sr) to improve the frequency stability.
J. Lodewyck — Sr OLCs – RTG April 2015
34/34