Transcript Grace`s DAMOP Slides - University of California, Berkeley

University of California, Berkeley
G. D. Chern
A. T. Nguyen
D. Budker
M. Zolotorev
http://phylabs.berkeley.edu/budker
1
•Motivation
•Population Scheme
•Adiabatic Passage
•Calculations
•Experimental Setup
•Data
•Experimental Results
•Conclusions
2
SEARCH FOR PARITY
NONCONSERVATON
(PNC) IN ATOMIC
DYSPROSIUM
3
PARITY TRANSFORMATION
•Mirror reflection
•Parity operator P and eigenstate Y(r)
P Y(r)= + Y(r) 
P Y(r)= - Y(r)
Y(r) EVEN

Y(r) ODD
•Interaction CONSERVES PARITY if
Y(r) simultaneously an
eigenstate of P and Hamiltonian H
4
Parity Nonconservation (PNC) in
Atoms
•CONSERVES PARITY:
Dominant electromagnetic interactions
e-
g
e-
q
q
•Definite parity eigenstates
Yeven , Yodd
5
•DOES NOT CONSERVE PARITY:
Smaller weak interactions between valence
electron and nucleus
e-
Z0
e-
q
q
•New eigenstates, e.g.
Yeven 
Yodd H w Yeven
Eeven  Eodd
6
Yodd
Why Dysprosium (Dy)?
•Pair of nearly-degenerate opposite parity
states
 enhances mixing
Energy (cm-1)
20,000
0
A
tA=7.9ms
B
tB >200ms
3.1 MHz
Ground State
•Isotopic comparisons:
Cancel Atomic Theory
•Odd isotopes:
Anapole moment
7
Other Applications for Dy
•Investigation of Sokolov effect
[B. B. Kadomtsev et al., Physica Scripta. 54, 156-162 (1996)]
•New PNC effects
 Hw
[T. Gasenzer et al., European Journal of Phys. (1999)]
8
•Atomic calculations predicted:
Hw=70  40 Hz
[V. A. Dzuba et al., Phys. Rev. A 50, 3812 (1994)]
•Using pulsed lasers, we reported:
|Hw|=|2  3| Hz
[A. T. Nguyen et al., Phys. Rev. A 56, 3453 (1997)]
•Improve statistical sensitivity
9
4f9 6s2 6p f
J=9
1397 nm (.30(9) b.r.)
4f10 5d 6s2 A
J=10
B 4f9 5d2 6s
J=10
669 nm
e 4f9 5d 6s2
J=8
833 nm
4f10 6s2 G
J=8
EVEN
ODD
10
•Send atomic beam across cw laser
laser beam
atomic
beam
•Consider a two-level atom
|e>
light
w0
atomic
states
|g>
•Oscillations at Rabi freq. WR = d E
 50% prob. to be in excited state
Prob. to be
in state |e>
1
time
11
•PROBLEM: cw laser excites only small
distribution
# of atoms
in state |g>
fraction of transverse atomic velocity
vT
•SOLUTION: diverge laser beam to match
atomic beam divergence
[C. R. Ekstrom et al., Optics Comm. 123.505 (1996)]
diverging laser beam
atomic
beam
•Due to Doppler effect, almost entire
distribution excited
12
•ADIABATIC PASSAGE: robust population
technique
 Allows 100% prob. to be in excited state
Prob. to be
in state |e>
1
time
•Each atom “sees” change in detuning as it
traverses laser beam
•Analogous to adiabatic passage in NMR
13
Dressed Atom Model
•Include atoms and light field in basis states
•“Dress” atomic states w/ photon states: |g,
n> and |e, n-1>
•Eigenstates in dressed atom basis:
|1> and |2>
Energy
|1>
|g, n>
|e, n-1>
Detuning
|2>
14
Adiabatic Criteria
•Change in detuning D is slow compared to
Rabi frequency WR
dD/dt  | W |
R
|W R |
•Lifetime of upper state t is longer than time
T for inversion to occur
t  T 
15
WR
dD/dt
•Hamiltonian:
d E (t ) 
 0

H  
 d E (t ) D(t ) 
•Density Matrix:
 11 12 


 21 22 
•Liouville Equation:
d
i
1
  [ H , ]  (   )
dt

2
16
OV
3
83
EN
c
PMT
e
b
a
c
d
6
-n
69
m
lig
ht
a) atomic beam produced by
effusive oven source at T=1500 K
b) atomic beam collimators
c) cylindrical lenses to diverge laser beams
d) spherical mirror to improve light collection
efficiency
e) interference filter(s)
17
- nm
lig
ht
•First transition: state G  state e
•833-nm fluorescence
18
•Probe with 669-nm laser
1
.
0
0
.
9
0
.
8
0
.
7
0
.
6
0
.
5
0
.
4
743-nmFluoresc(ab.units)
0
.
3
0
.
2
0
.
1
0
.
0
1
0
0
8
0
6
0
4
0
2
0
0
2
0
4
0
6
0
8
0
1
0
0
6
6
9
n
m
D
e
t
u
n
i
n
g
(
M
H
z
)
19
•FIRST transition: ~50% efficiency
•Limited only by insufficient laser power
 need to double power
•SECOND transition: ~80% efficiency
•THIRD transition: 30% efficiency
•determined by branching ratio
20
TOTAL EFFICIENCY:
(~50%) * (~80%) * (30%) = ~12%
without lenses: < 0.5% efficiency
•Efficiency easily improved w/:
•more 833-nm laser power
•1397-nm laser to stimulate
transition
21
•Pulsed lasers  cw lasers:
104 increase in duty cycle
•With 20 h data taking time, this gives
total factor of 102 increase in
statistical sensitivity (10 mHz level)
22
Stark-PNC Mixing
A
B
+
Stark Mixing
Stark-PNC Mixing
=
23