one, two, three, many creating quantum systems one atom at ...Β Β· β’ extremely low momentum, such...
TRANSCRIPT
Selim Jochim, UniversitΓ€t Heidelberg
One, two, three, many
Creating quantum systems one atom at a time
Selim Jochim, UniversitΓ€t Heidelberg
One, two, three, many
Creating quantum systems one atom at a time
Selim Jochim, UniversitΓ€t Heidelberg
One, two, three, many
Creating quantum systems one atom at a time
Two trapped interacting particles β¦
interacting singlet
Ground state of the Helium atom:
No analytic solution available, we learn how
to apply powerful numerical techniques:
Hartree Fock method.
Many body limit
Define quantities like the Fermi energy,
density, pressure β¦.
β¦ apply local density approximation β¦
But when are such approximations
justified?
This is an ancient problem!
Sorites Paradox
How many grains make a heap?
β’ 1 grain of sand does not make a heap.
β’ If 1 grain does not make a heap then 2 grains of sand do not.
β’ If 2 grains do not make a heap then 3 grains do not.
β’ β¦
β’ If 9,999 grains do not make a heap then 10,000 do not.
From Stanford Encyclopedia of Philosophy:
http://plato.stanford.edu/entries/sorites-paradox/
http://commons.wikimedia.org/wiki/File%3ASossusvlei_Dune_Namib_Desert_Namibia_Luca_Galuzzi_2004.JPG
Ultracold neutral atoms
Bose Einstein condensates of large samples of atoms: Macroscopic wave
function: Number of particles is so large that a constant density of atoms is
observed in experiments:
http://jila.colorado.edu/bec/images/bec.png
Removing one single atom does not make a difference!
Measure: π π = β¨ Ξ¨β π Ξ¨ π β©
Our approach
Reduce the complexity of a system as much as possible
until only the essential parts remain!
In most physical systems:
Range of interaction
significantly complicates the description
Ultracold atoms are an ideal tool β¦
The interactions between ultracold atoms can be effectively pointlike
(contact interaction)
van der Waals interaction: range of ππ£ππ βΌ 1nm
In the experiments we have:
β’ extremely low density (interparticle spacing ~ 1Β΅m)
β’ extremely low momentum, such that πππ΅ =β
2ππππβ« ππ£ππ
β’ extremely low momentum, such that πππ΅ =β
2ππππβ« ππ£ππ
(This is the opposite limit desired in collision experiments:
shorter wavelength enhances resolution)
Here:
β’ If πππ΅ is sufficiently large, all the information about internal structure of the
atom is hidden in a single quantity, the scattering length π
β’ We can even tune the scattering length to any desired value by simply
applying a magnetic field (Feshbach resonances).
Ultracold atoms are an ideal tool β¦
The 6Li atom
11
0 200 400 600 800 1000 1200-5
0
5
Scatt
eri
ng
len
gth
(1000 a
0)
Magnetic field (G)
S=1/2, I=1
β half-integer total angular momentum
β 6Li is a fermion
F=3/2
En
erg
y
magnetic field [G]
F=1/2
|>|>
mI= 0
mI= 1
Tuning interactions: Feshbach resonance in 6Li
aββ
6Li ground state
NO interaction between identical particles
C. Regal et al., Nature 424, 47-50 (2003)
Two-body system: Tune the binding
energy of a weakly bound molecule:
Tunability of ultracold systems
600 800 1000 1200-10
-5
0
5
10
Scatt
eri
ng
len
gth
(1000 a
0)
Magnetic field (G)
Binding energy:
Size (>> range of interaction):
Feshbach resonance: Magnetic-
field dependence of s-wave
scattering length
/( ) r ar e
2
2BEma
We have prepared such molecules up to ~1Β΅m in size!
G. ZΓΌrn et al., PRL 110, 135301 (2013)
This talk
Outline
β’ How do we prepare our samples?
β’ How many particles do we need to form a heap?
β’ Controlling the motion of two particles in a double well
A picture from the lab β¦
109 laser cooled atoms at ~1mK
Thomas Lompe ( MIT)
Matthias Kohnen
( PTB Braunschweig) Friedhelm Serwane
( UC S. Barbara) Timo
Ottenstein
( Dioptic)
A container for ultracold atoms
This might still work
for liquid nitrogen β¦.
We need to isolate the atoms from the environment:
β¦ here we use the focus of a laser beam:
Optical dipole trap depth: π β πΌ(π)
Evaporative cooling
It just works the same:
For our cup of coffee β¦
β¦ and for our cold atoms:
Cool from ~1mK down to
below 1Β΅K
Just reduce the trap depth, i.e. laser power
Ultracold gas of fermions
About 50000 atoms @ 250nK, TF~1Β΅K
~100Β΅m
Absorption imaging of ultracold clouds:
resonant laser beam
CCD camera
Single atom detection
CCD
1-10 atoms can be distinguished with
high fidelity > 99%
one atom in a MOT
1/e-lifetime: 250s
Exposure time 0.5s
Fluorescence normalized to atom number
Towards a finite gas β¦
The challenge:
achieve βΟ β« ππ
Creating a finite gas of fermions
β’ superimpose microtrap (~1.8 Β΅m waist)
p0= 0.9999
β’ 2-component mixture in reservoir
E
n
1
Fermi-Dirac dist.
~100Β΅m
Creating a finite gas of fermions
β’ switch off reservoir
p0= 0.9999
+ magnetic field gradient in
axial direction
Spilling the atoms β¦.
β’ We can control the atom number with exceptional precision!
β’ Note aspect ratio 1:10: 1-D situation
β’ So far: Interactions tuned to zero β¦
0 1 2 3 40
10
20
30
40
50
60
70
80
90
100
2% 2%
co
un
ts
fluorescence signal
96%
5 6 7 8 9 100
20
40
60
80
100
120
140
6.5%5%
88.5%
co
un
ts
fluorescence signal
F. Serwane et al., Science 332, 336 (2011)
Letβs study the interacting
system!
Precise energy measurements
βbareβ RF β transition
RF β transition with interaction
RF photon+ ΞE
RF photon
Radio Frequency spectroscopy
Measure the interaction energy
vary the number of majority particles:
0,0 0,5 1,0 1,5 2,0 2,5 3,00,0
0,2
0,4
0,6
interaction shift [kHz]
Tra
nsfe
rre
d f
ractio
n
0 1 2 3 4 5
0,0
0,5
1,0
1,5
2,0
E
[Δ§
ll]
N
Measure the interaction energy
vary the number of majority particles:
0,0 0,5 1,0 1,5 2,0 2,5 3,00,0
0,2
0,4
0,6
interaction shift [kHz]
Tra
nsfe
rre
d f
ractio
n
0 1 2 3 4 5
0
1
2
g
=0.25
g
=0.8
g
=2.0
E
[Δ§
ll]
N
10 8 6 4 2 0
0,07
0,1
0,4
0,7
1
kF/g
E/E
F
lnteraction energy in dimensionless units
Analytic solution of the two particle problem
T.Busch et al., Found. Phys. 28, 549 (1998)
Analytic solution for an infinite number of majority particles
J. McGuire, J. Math. Phys. 6,432 (1965)
(local density approximation)
noninteracting strongly repulsive
A. Wenz et al., Science 342, 457 (2013)
0 1 2 3 4 5
0
1
2
g
=0.25
g
=0.8
g
=2.0
E
[Δ§
ll]
N
10 8 6 4 2 0
0,07
0,1
0,4
0,7
1
kF/g
E/E
F
lnteraction energy in dimensionless units
Nβ=1
Nβ=2Nβ=3Nβ=4Nβ=5
Analytic solution of the two particle problem
T.Busch et al., Found. Phys. 28, 549 (1998)
Analytic solution for an infinite number of majority particles
J. McGuire, J. Math. Phys. 6,432 (1965)
(local density approximation)
noninteracting strongly repulsive
A. Wenz et al., Science 342, 457 (2013)
10 8 6 4 2 0
-0,01
0,00
0,01
0,02
0,03
0,04
0,05
-E
/EF
-kF/g
0,07
0,1
0,4
0,7
1
E/E
F
lnteraction energy in dimensionless units
Nβ=1
Nβ=2Nβ=3Nβ=4Nβ=5
2 particles
N particles
3 particles
S.E. Gharashi et al.,
PRA 86, 042702 (2012)
A. Wenz et al., Science 342, 457 (2013)
G.E. Astrakharchik and
I. Brouzos, arxiv:1303.7007
Approaching the many body limitβ¦
β¦ with very few particles (in a one-dimensional system)
A. Wenz et al., Science 342, 457 (2013)
Interesting things to look at:
β’ Polaron physics in various dimensions
β’ The Kondo problem
β’ Andersonβs orthogonality catastrophe
Realize multiple wells β¦
β¦.. with similar fidelity and control?
Basic building blocks of matter!
The setup
Light intensity distribution
A tunable double well
J
initial spatial wave function:
spin wave function (stationary)
β’ Interactions switched off:
A tunable double well
J
0 25 50 750
1
2
Ato
m n
um
ber
in w
ell
|R>
Time (ms)
well |π β©well |πΏβ©
switch off left
well before
counting
atoms
Two interacting atoms
U
J
0 25 50 750
1
2
Ato
m n
um
be
r in
we
ll |R
>
Time (ms)
c)
well |π β©well |πΏβ©
Interaction leads to entanglement:
Two strongly interacting atoms
U
J
well |π β©well |πΏβ©
U
J
well |π β©well |πΏβ©
In a balanced double well, they can only tunnel together!
Two strongly interacting atoms
U
J
well |π β©well |πΏβ©
U
J
well |π β©well |πΏβ©
We can compensate for the interaction energy by applying a tilt!
Two strongly interacting atoms
β’ Observe number statistics in the right well (time averaged)
-15 -10 -5 0
0,0
0,5
1,0
2 atoms
Pro
ba
bili
ty in
we
ll |R
>
tilt [J]B=740G
two atoms
Two strongly interacting atoms
β’ Observe number statistics in the right well (time averaged)
-15 -10 -5 0
0,0
0,5
1,0
1 atom
2 atoms
Pro
ba
bili
ty in
we
ll |R
>
tilt [J]B=740G
U
two atoms
one atom
Two strongly interacting atoms
β’ Observe number statistics in the right well (time averaged)
-15 -10 -5 0
0,0
0,5
1,0
0 atoms
1 atom
2 atoms
Pro
ba
bili
ty in
we
ll |R
>
tilt [J]B=740G
two atoms
one atom
no atomsU
Preparing stationary states
β’ If we ramp on the second well slowly enough, the system will remain in its
ground state:
Eigenstates of a symmetric DW
1
2(| β©πΏπΏ + | β©π π )
1
2(| β©πΏπ + | β©π πΏ )
1
2(| β©πΏπΏ + | β©πΏπ + | β©π πΏ + | β©π π )
Preparing stationary states
β’ Number statistics for the balanced case depending on the interaction
strength:
-2 0 2 4 6 8 100,0
0,5
1,0
Pro
ba
bili
ty
Interaction U (J)
Single occupancy
Preparing stationary states
β’ Number statistics for the balanced case depending on the interaction
strength:
-2 0 2 4 6 8 100,0
0,5
1,0
Pro
ba
bili
ty
Interaction U (J)
Single occupancy
Double occupancy
Measuring energies
13
Trap modulation spectroscopy
Measuring energies
First-order
tunneling ~ 2π½ Second-order
tunneling ~4 π½
π
2
13
Trap modulation spectroscopy
Measuring energies
First-order
tunneling ~ 2π½ Second-order
tunneling ~4 π½
π
2
Super exchange energy!
responsible for spin ordering in the many body ground state
13
How to go to a many-body system?
β’ Inspired by a top-down approach: D. Greif et al., Science 340, 1307-1310 (2013) (ETH ZΓΌrich)
β’ Dimerize a lattice filled with spin-1/2 fermions to observe spin correlations
Outlook
Combination of multiple double wells
β’ Preparation of ground states
in separated double wells
β’ Combination to larger system
Can this process be done adiabatically ?
Can it be extended to larger systems ?
14
Vincent KlinkhamerGerhard ZΓΌrn
Thank you for your attention!
Funding:
AndreWenz
Thomas Lompe(-> MIT)
DhruvKedar
SelimJochim