Philip Harris

1
Testing Time Reversal the search for the neutron
electric dipole moment

• Philip Harris

2
Overview
1. nEDM Classic

• New limit dn lt 2.9 x 10-26 e.cm
• Discovery of new systematic effect

2. CryoEDM at ILL

• Starting shortly 100x improved sensitivity

3. Other EDM searches
3
Electric Dipole Moments

• EDMs are P, T odd
• Complementary study of CPv
• SM CPv parameterised, notunderstood
• Important to test with other systems than K, B
• Clean system
• background free SM predicts tiny EDMs, other
  models typically 106 larger
• Tight constraints on models of new physics

4
CPv and the Baryon Asymmetry

• A.D. Sakharov, JETP Lett. 5, 24-27, 1967
• Baryon number violation
• Not allowed at tree level, but permitted in
  higher-order processes in SM
• Departure from thermal equilibrium
• Expansion of Universe
• Phase transitions
• CP violation
• Note SM CPv is orders of magnitude too small to
  explain observed asymmetry we need new physics.
  
• Larger-than-SM CPv tends to predict
  larger-than-SM EDMs

5
Standard Model EDM

• EDMs are zero-momentum limit of
  fermion-fermion-photon 3-point function
• SM CPv related to single phase in CKM, ? flavour
  change
• EDM is flavour conserving CPv, ? flavour change
  squared
• In fact ??3 loops v. suppressed

6
Strong CP problem
CPv phase qQCD in strong int. induces neutron EDM
dn 10-16 q e.cm ? q lt 2?10-10 rads

• Why is q so small?
• Peccei-Quinn Axions?

7
Supersymmetry

• SUSY breaking introduces new CPv phases
• EDMs appear at 1-loop level
• For superpartner masses few 100 GeV large
  phases, predicted nEDM is 100x too large SUSY
  CP problem

8
SUSY constraints example
MSUSY 500 GeV tan b 3
Pospelov Ritz, hep-ph/0504231
9
SUSY, contd
Lebedev et al., hep-ph/0402023
10
EDM candidates

• Neutron Intrinsic EDM from CPv field thy (SUSY,
  L-R, addl Higgs...) strong CPv
• Electron Intrinsic EDM from CPv field thy
• Atomic P, T violating nucleon-nucleon or
  nucleon-electron interaction (inc. strong CPv)
• all sensitive to different physics beyond SM
• each gives new window on CPv
• must measure all to distinguish between CPv models

11
History

• Factor 10
• every 8 years
• on average

12
Reality check

• If neutron were the size of the Earth...

13
Measurement principle
Use NMR on ultracold neutrons in B, E fields.
B0
B0
B0
E
E
ltSzgt h/2
h?(0)
h?(??)
h?(??)
ltSzgt - h/2

• ?(??) ?(??) 4 E d/ h
• assuming B unchanged when E is reversed.

Energy resolution of our detector
lt10-21 eV
14
ILL, Grenoble
15
The ILL reactor
16
Apparatus
Magnetic shielding
Storage cell
N
S
Magnet polarizing foil
Ultracold neutrons (UCN)
UCN detector
17
Apparatus
HV feedthru
Neutron storage chamber
B-field coils
18
Measurement principle
Measure Larmor spin precession freq in parallel
antiparallel B and E fields
B
E
mB
Reverse E relative to B, look for freq shift.
19
Ramsey method of Separated Oscillating Fields
Spin up neutron...
1.
Apply ?/2 spin-flip pulse...
2.
Free precession...
3.
Second ?/2 spin-flip pulse
4.
20
Ramsey resonance

• 2-slit interference pattern
• Phase gives freq offset from resonance

21
nEDM measurement
22
Hg co-magnetometer
dHg lt 2.1 x 10-28 e cm Romalis et al., PRL 86
(2001) 8505
Polarised Hg atoms
PMT
PMT output
23
Mercury frequency
24
nEDM measurement
25
Neutron EDM results (binned)
...but B up, B down disagree.
26
Leading systematic
Combination of two effects
27
Geometric Phase
J. Pendlebury et al., PRA 70 032102 (2004) P.
Harris, J. Pendlebury, PRA 73 014101 (2006)
... so particle sees additional rotating field
Bottle (top view)
Frequency shift ? E
Looks like an EDM, but scales with dB/dz
28
Geometric Phase How to measure it

• Consider
• Should have value 1
• R is shifted by magnetic field gradients
• Plot EDM vs measured R-1

29
Geometric Phase
Magnetic field down
30
Geometric Phase
Magnetic field up
31
Results
EDM
R-1
0
Nearly...
32
Results
Small dipole/quadrupole fields ( Earths
rotation!) can pull lines apart add GP shifts
EDM
B up
R-1
0
B down
33
Results
EDM
B up
R-1
0

• Use variable-height bottle to measure B field
  shape
• Depolarization data help to establish
  separations
• Apply corrections

B down
34
Final Result
Latest limit dn lt 2.9 x 10-26 e.cm (90 CL)
C.A. Baker et al., Phys. Rev. Lett. 97, 131801
(2006), hep-ex/0602020
35
CryoEDM The Next Generation
New technology

• More neutrons
• Higher E field
• Better polarisation
• Longer NMR coherence time

• 100-fold improvement in sensitivity

36
UCN production in liquid helium
R. Golub and J.M. Pendlebury Phys. Lett. 53A
(1975), Phys. Lett. 62A (1977)

• 1.03 meV (11 K) neutrons downscatter by emission
  of phonon in liquid helium at 0.5 K
• Upscattering suppressed Boltzmann factor e-E/kT
  means not many 11 K phonons present

37
UCN production rate vs ln
1.19?0.18 UCN cm-3 s-1 expected, 0.91? 0.13
observed C.A.Baker et al., Phys.Lett. A308 67-74
(2002)
38
CryoEDM overview
Neutron beam input
Cryogenic Ramsey chamber
Transfer section
39
Cryogenic Ramsey chamber
HV electrode
HV feed
SQUID loops (not shown)
Superfluid He
n storage cells (4 eventually)
40
UCN detection in liquid helium

• Solid-state detectors developed for use in LHe
• Thin surface film of 6LiF n 6Li ? a 3H

C.A.Baker et al., NIM A487 511-520 (2002)
41
Detection of polarised UCN

• Downscattered UCN remain polarised first
  observation Sept 2007

42
Ramsey cell and HV stack
43
Statistical limits
h / 2
?d
? E T ?N
Parameter Room-tmpr. expt Sensitivity

• Polarisationdetection ? 0.75 x 1.2
• Electric field E 106 V/m x 4
• Precession period T 130 s x 2
• Neutrons counted N 6 x 106 /day x 4.5
• (with new beamline) (x 6) x 2.6

Total increase approx factor 100
44
Systematics

• B-field fluctuations
• Superconducting solenoid shield will give much
  improved field shape and shielding
• SQUIDs give temporal variation (common-mode)
• Shielding not optimal at present improvements
  foreseen
• Geometric phase
• n are 40x less sensitive than Hg
• 1/B2 5 x increase in B gives 25x protection
• overall 1000x improvement

45
Systematics

• Other
• Exv 3E-29 e.cm
• Feedback from field coils lt1E-30 e.cm
• Electrostatic forces lt 1E-28 e.cm
• Leakage currents 1 nA ? 5E-29 e.cm
• AC fields from HV lt 1E-29 e.cm
• Overall no show-stoppers...

46
Current status

• Commissioning underway
• UCN and magnetometry tests continuing
• B shielding not yet optimal, and electric field
  needs to increase
• First results 2009 at 1E-27 level
• New beamline 2012?

47
Other nEDM expts SNS, ORNL
Concept by Golub Lamoreaux, 1994
Large US collaboration 40 people, 16M
Anticipated sensitivity lt 10-28 e.cm
Construction complete 2013 at earliest
48
Other nEDM experiments PSI

• spallation target
• D2O moderator
• currently testing apparatus at ILL
• move to PSI inJan 09

49
Other nEDM experiments PNPI/ILL

• Serebrov group tried to build multi-chamber
  spectrometer. B-field quality not adequate
• Currently installing upgraded apparatus from old
  PNPI expt (1990)
• Hopes to start running in 2009
• Sensitivity 10-26 e cm after 1 yr

50
Electron EDM Berkeley

• Unpaired e- inneutral atom
• Pairs of Tl beams in opposite E fields
• Na beams as comagnetometer

Final result delt1.6 x 10-27 e.cm (systematics
limited)
51
Imperial College (Hinds) eEDM expt
Part of optical setup.... Unconventional
particle physics!

• Current status
• Stat sensitivity beats current limit
• Systematics under investigation

52
Electron EDM measurements
Group System Advantages Projected gain
D. Weiss (Penn St.) Trapped Cs Long coherence 100!
D. Heinzen (Texas) Trapped Cs Long coherence 100!
H. Gould (LBL) Cs fountain Long coherence ?
L. Hunter (Amherst) GdIG solid Huge S/N 100?
S. Lamoreaux (LANL) C.-Y. Liu (Indiana) GGG solid Huge S/N 100?-100,000?
E. Hinds (Imperial) YbF beam Large Internal E 3, then 30
D. DeMille (Yale) PbO cell Int.Egood S/N 30!-1,000?
E. Cornell (JILA) trapped HBr Int. E long T ??
N. Shafer-Ray (Okla.) trapped PbF Int. E long T ??
(This slide mainly from DeMille, PANIC 05)
53
199Hg EDM

• Optically pumped 199Hg atoms precess in B, E
  fields, modulating absorption signal

• Dual cells remove effect of drifts in B

• Result d(199Hg) lt 2.1 x 10-28 e cm
• Provides good limit on CPv effects in nuclear
  forces, inc. qQCD
• If from valence neutron, corresponds to
  dnlt2x10-25 ecm, because of electrostatic
  shielding.

54
... and more!

• Muon EDM 7x10-19 e.cm from g-2 (New proposal
  at JPARC 10-24)
• Deuterium EDM similarprinciple 10-29
  achievable?
• Tau weak dipole moment CP-odd observables in
  diff. x-sec at Z res.,6E-18 e.cm from LEP data

Sensitivity to physics BSM depends on source of
CPv
55
Conclusions

• New nEDM limit, 2.9 x 10-26 e.cm
• Systematics understood as never before
• CryoEDM coming soon 100x more sensitive
• Competition on its way
• Watch this space!

56
And finally...

• It may be that the next
• exciting thing to come along will be the
  discovery of a neutron or
  atomic or electron EDM. These EDMs...
  seem to me to offer one of the most
  exciting possibilities for progress in particle
  physics.
• - S. Weinberg

57
Further reading

• Web site www.neutronedm.org
• Room-temperature result hep-ex/0602020
• Theory see, e.g., refs in hep-ex/0602020
• Geometric phase J. Pendlebury et al., PRA 70
  032102 (2004) and P. Harris, J. Pendlebury, PRA
  73 014101 (2006)
• n production in helium C.A.Baker et al.,
  Phys.Lett. A308 67-74 (2002)
• Detectors C.A.Baker et al., NIM A487 511-520
  (2002)