Electric dipole moments: theory and experiment
EA Hinds
Blois June 2002
Two motivations to measure EDMs
EDM is effectively zero in standard modelbut
big enough to measure in non-standard models
direct test of physics beyond the standard model
EDM violates T symmetry
Deeply connected to CP violation and the matter-antimatter asymmetry of the universe
φ ∼ 1
Left-Right
10-32
10-20
10-22
10-24
10-30
MultiHiggs SUSY
φ ∼ α/π
Standard Model
Electro-magnetic
10-34
10-36
10-38
Exp
erim
enta
l Lim
it on
d (
e.cm
)
1960 1970 1980 1990
neutron:electron:
2000
10-20
10-30
10-22
10-24
10-26
10-28
A bit of history
thallium
atom/moleculelevel
CP from particles to atoms (main connections)
nuclearlevel
NNNNSchiff
momentmercury
HiggsSUSY
Left/Right
StrongCP
field theoryCP model
θGG%
neutron
nucleonlevel
electron/quarklevel
de
dq
dcq
The Mercury EDM experimentUniversity of Washington, Seattle
M.V. Romalis, W.C. Griffith, J.P. Jacobs, E.N. Fortson
Nuclear spin polarised 199Hg vapor in a double cell
hω = µB
+ dEE
- dE±
B
ω± measured optically
199
Hg
ED
M (1
0-27
e.cm
)
GSI992
Hg edm result: Phys. Rev. Lett. 86, 2505 (2001)
|dHg| < 2.1×10-28 e cm
E = 9 kV/cm
B = 15 µT stable to 0.4 pT
Tcoherence ~ 100 s
Difference of precession frequencies gives 199Hg EDM
The Neutron EDM
Rutherford-Appleton LaboratoryCA Baker, K Green, P Iaydjiev, S Ivanov
Sussex UniversityS Al-Ayoubi, PG Harris, JM Pendlebury, JD Richardson, D Shiers, K F Smith, M van der Grinten
ILLP Geltenbort
• Ultracold neutrons in a bottleprecess at frequency (µnB ± dnE)/h
• Reverse E to measure dn
GSI992
Neutron edm result: Phys. Rev. Lett. 82, 904 (1999)
|dn|< 6.3×10-26 e cm
E = 4.5 kV/cm
B = 1 µT
Tcoherence ~ 130 s
Hg co-magnetometer
(B is measured to 1 pT rms)
Aiming at 10-27 e.cm over next decade
Neutron: longer range plans
§ Helium (ILL, LANL)Inside the heliumHigher neutron densityHigher E field
New moderators using liquid helium or solid deuterium
§ Deuterium (PSI, TU-Munich)Outside the deuteriumHigher neutron densityBigger volume (fast moderation)
Implications of n and Hg for the theta parameter
θ gs2
32π2GG~
CP strong interaction
dn» 10-16 θ
\ θ < 6×10-10n n
γ
p
π −
×
induces neutron EDMBaluniCrewtherPospelov
Henley & HaxtonPospelov
induces mercury Schiff moment
N/
π ×N
dHg» 3 10-19
θ
\ θ < 6×10-10
Something(Peccei-Quinn?)
makes θ very small!
Implications of Hg and nfor SUSY
quark electric dipole moments
dq q (Fµν σµνiγ5 ) q21
q q
γ
gaugino
squark
quark color dipole moments
q q
g
gaugino
squark
dq q (gs Gµνσµνiγ5 ) q21 c
2Λmqdq, dq ~ (loop factor) × sin ϕCP
cCP phase from soft breaking
naturally O(1)
scale of SUSY breaking naturally ~200 GeV
naturally ~ α/π
du,d, du,d ~ 1×10−23 cm naturallyc
n and Hg experiments give du< 2 × 10−25
dd< 5 × 10−26
du< 3 × 10−26
dd< 3 × 10−26
c
c
~ 300 times less!
ϕCP < 3×10-3 ??
Λ > 5 TeV ??
thallium
atom/moleculelevel
CP from particles to atoms (main connections)
nuclearlevel
NNNNSchiff
momentmercury
HiggsSUSY
Left/Right
StrongCP
field theoryCP model
θGG%
neutron
nucleonlevel
electron/quarklevel
de
dq
dcq
And now for the electron………..
The Thallium EDM experimentBerkeley B.C. Regan, E.D. Commins, C.J. Schmidt and D. DeMille
2 Tl atomic beams
hω = µB
polarise
analyse
± dEE±B
The solution:add 2 more Tl beams going down
4
analyse
polarise
The solution:Add 4 Na beams for magnetometry
1st huge problem:motional interaction µ • v × E
2nd huge problem:stray static magnetic fields
A beautiful feature of the method
E
electric field
Interaction energy
-de ηE•σ
ηde σ
atom containing electron
amplification
-585 for Tl
(Sandars)
Effective field = 72 MV/cm× 585
÷ 585electron edm result:
|de|< 1.6 ×10-27 e.cm
E = 123 kV/cm
B = 38 µT
Tcoherence = 2.4 ms
Na co-magnetometer
Final Tl result: PRL 88, 071805 (2002)
|dTl|< 9.4 ×10-25 e.cm
Theoretical consequences of electron EDM
SUSY electron edm
e e
γselectron2Λ
mede ~ (loop) × sin ϕCP
No direct contamination from θ problem
- a pure new physics search
~ 5 × 10−24 cm
Once again, natural SUSY is too big by 300
ϕCP < 3×10-3 ?? Λ > 5 TeV ??
potentially 1000 × more sensitive
The Sussex experiment uses ytterbium fluoride moleculesJJ Hudson, BE Sauer, MR Tarbutt and EA Hinds,
arXiv hep-ex0202014 (2002)
The future for electron EDM experiments
polar molecules
0
5
10
15
20
0 10 20 30
Applied field E (kV/cm)
Effe
ctiv
e fie
ld η
E (G
V/c
m)
(in Tl experiment ηΕ was 72 MV/cm)
13 GV/cm
First advantage of YbF: Huge effective field ηEParpiaQuineyKozlovTitov
2nd advantage of YbF:No coupling σ × E • v to motional magnetic field
and internuclear axis is coupled to E
E
∴< σ × E > = 0 no motional systematic error
electron spin σ is coupled to internuclear axis
σ
Yb+
F-
B and E1.5 m
The Sussex molecular beam
beam
oven
state select
polarise
analyse
detect
PMT
Part of the optical setup
Measuring the edm
Applied magnetic field
Det
ecto
r co
unt r
ate
cos2 ω+T
B0
cos2 ω−T
δφ = 4deηET/h
-B0
− 4deηET/h
Projections for the future
background
fringe height
coherence time
de in 1 day
2002result
150kHz
1.5 kHz
1.5 ms
3 10-26 e cm
cold YbFbeam
640kHz
160 kHz
1 ms
6 10-28 e cm
trappedmolecules
40kHz
10 kHz
1 s
3 10-30 e cm
long time = narrow fringes
d(muon) < 7�10-19
φ ∼ 1
Left-Right
10-20
10-22
10-24
d e.cm
MultiHiggs SUSY
φ ∼ α/π
Electro-magnetic
neutron:electron:
1960 1970 1980 1990 2000 2010 2020 2030
10-28
10-29
Current status of EDMs
d(neutron) < 6�10-
26
d(proton) < 6�10-
23YbF expt
coldmolecules
d(electron) < 1.6�10-
27
Conclusion
EDM measurements (especially cold moleules)have great potential to elucidate
• CP violation
• particle physics beyond the standard model
• matter/antimatter asymmetry of the universe
some of the most fundamental issues in physics