Strongly Correlated Materials made out of Ultra Cold Atoms Tin-Lun - - PowerPoint PPT Presentation

strongly correlated materials made out of ultra cold atoms
SMART_READER_LITE
LIVE PREVIEW

Strongly Correlated Materials made out of Ultra Cold Atoms Tin-Lun - - PowerPoint PPT Presentation

Strongly Correlated Materials made out of Ultra Cold Atoms Tin-Lun Ho The Ohio State University PSM2010 Yokohama, Japan, March 11, 2010 Ho and Zhou, PNAS 2009 (cooling scheme using band insulator) Ho and Zhou, Nature physics 2010


slide-1
SLIDE 1

Strongly Correlated “Materials” made out of Ultra Cold Atoms

Tin-Lun Ho The Ohio State University PSM2010 Yokohama, Japan, March 11, 2010

slide-2
SLIDE 2

Ho and Zhou, PNAS 2009 (cooling scheme using band insulator) Ho and Zhou, Nature physics 2010 (deducing bulk properties from trap data) Zhou and Ho: universal thermometry (cond-mat/09) Ho and Zhou: Universal cooling scheme (cond-mat/09) Ho and Li: Quantum Hall Needles in Synthetic Gauge Fields (to be published)

slide-3
SLIDE 3

Great interests in strongly correlated soon after discovery of BEC Strongly Interacting Fermi (and Bose) gases Bosonic Quantum Hall states Optical Lattice Emulator Spin-1 Boson singlet Low D quantum Gases, disordered quantum gases Systems being studied in various labs at present : ENS, Paris Expt: Munich, MIT, Rice, ETH, NIST, UIUC, Penn State Diploar gases

slide-4
SLIDE 4

Quantum Simulation: Most ambitious project in cold atoms ever

* To find solutions to unsolved problems/models * As a calibration for theories.

slide-5
SLIDE 5

Main Challenge in QS Ultra low T Very low S/N

Very small energy scales in this regime

1012 1014K

How to obtain information of bulk systems from Data of trapped gases ?

Quantum Many-body precision measurement

slide-6
SLIDE 6

Part I: Quantum Simulation via Optical Lattice Emulator

slide-7
SLIDE 7

Optical lattice Produced by a pair of counter propagating laser

slide-8
SLIDE 8
slide-9
SLIDE 9
slide-10
SLIDE 10

Figure 2 Absorption images of multiple matter wave interference patterns. These were obtained after suddenly releasing the atoms from an optical lattice potential with different potential depths V0 after a time of flight of 15 ms. Values of V0 were: a, 0 Er; b, 3 Er; c, 7 Er; d, 10 Er; e, 13 Er; f, 14 Er; g, 16 Er; and h, 20 Er.

  • M. Greiner et.al, Nature 415, 39 (2002)
  • M. Greiner, O. Mandel. Theodor, W. Hansch & I. Bloch,Nature (2002)

Observation of Superfluid-insulator transition superfluid Mott

slide-11
SLIDE 11

Typical density profile in an optical lattice : Wedding cake structure r n(r)

slide-12
SLIDE 12

Boson Relevant energy scales: Band gap Virtual hopping Strong correlation! Hopping Fermion

slide-13
SLIDE 13

87Rb 40K

slide-14
SLIDE 14

evaporation

Source of difficulty I: Conventional evaporation fails.

slide-15
SLIDE 15

Current methods of achieving strongly correlated states: Raising the optical lattice in a trapped gas:

slide-16
SLIDE 16

Spin disordered Mott insulator Even if evaporation works, there is another difficulty.

T  Tc 1kB /sec

slide-17
SLIDE 17

Part I : To realize the full power of quantum simulation

slide-18
SLIDE 18

Bulk thermodynamic properties of interest: Equation of state  phase boundary Entropy density Superfluid density Compressibility Spin susceptibility Staggered magnetization

n  n(,T) s  s(,T) s  s(,T) ˜ m  ˜ m (,T) T (,T)  n(,T) 

slide-19
SLIDE 19

Simplest example: Equation of state n  n(,T)

 n(r r )  n((r r ),T)  no( V (r r ),T)

Density of trapped gas Density of homogenous system If LDA works, then the experimental data immediately gives

 n(r r )

Local density approximation (LDA) :

 Q(r r )  Q( V (r r ))

n(,T)

  • LDA is valid for N>50 in typical traps
slide-20
SLIDE 20

Presence of phase transition: boundary:

dP  nd  sdT

PI (,T)  PII (,T)

First order transition: , discontinuous T I II

n s

n  s 

Continuous transition: change of slope in ,

r n,s

1st order Continuous transitions

 n(r r )

r (r)

slide-21
SLIDE 21

Use the tail of the density or number fluctuation At the surface, density is low, can do fugacity expansion

 n( x )  e( V ( 

x )) / T / 3

 ˜ n (x,y)   T       e( V ( 

r )) / T

2

column density Essential step of this procedure :

  • high quality density data
  • Accurate determination of and

T 

slide-22
SLIDE 22
  • III. Entropy Density

Need two configurations of different Compare them at the same

 '

T 

P(x,0,0)  P((x),T)

T' '

P'(x,0,0)  P('(x),T')

s  P T      

s  s(,T)  s  s( r ) dP  nd  sdT dP  nd

T 

slide-23
SLIDE 23

s(x,0,0)  P'(x',0,0)  P(x,0,0) T'T '(x')  (x)

(x)    1 2 M 2x2   ' 1 2 M'2 x'2  '(x')

slide-24
SLIDE 24

Superfluid density:

 dP  nd o  sdT Mns  w  d  w

  w   v

s  

v

n

 P  P(T,o,  w )

For a superfluid

n w2      

 o, T

  M 2 ns  o      

w2, T

  v

n  ˆ

z   x

Spatial changes in  changes in with respect to rotation

ns ns n

slide-25
SLIDE 25

Exploring the thermodynamics of a universal Fermi gas

  • S. Nascimbene, N. Navon, K. Jiang, F. Chevy, C. Salomon

arXiv: 0911.0747 Measurement of Universal Thermodynamic Functions for a Unitary Fermi Gas, Munekazu Horikoshi,1* Shuta Nakajima,2 Masahito Ueda,1,2 Takashi Mukaiyama1,3 Science, 442, vol 327, (2010)

3D Fermi gas (infinite Scattering length)

Apply the algorithm of Ho and Zhou

slide-26
SLIDE 26

Advantage of Cs in 2D lattice : low lattice depth, large lattice constant

In-situ Observation of Incompressible Mott-Insulating Domains of Ultracold Atomic Gases Nathan Gemelke, Xibo Zhang, Chen-Lung Hung, and Cheng Chin Nature 460, 995 (2009)

slide-27
SLIDE 27

Estimate of T based on number fluctuation, and Determination of phase diagram

T N   N 2    N 2

slide-28
SLIDE 28

Cheng Chin et.al

slide-29
SLIDE 29

Spin-Imbalance in a One-Dimensional Fermi Gas, Yean-an Liao, Ann Sophie C. Rittner, Tobias Paprotta, Wenhui Li, Guthrie B. Partridge, Randall G. Hulet, Stefan K. Baur & Erich J. Mueller arXiv.0912.0092

slide-30
SLIDE 30

arXiv: 0908.0174

slide-31
SLIDE 31
slide-32
SLIDE 32
slide-33
SLIDE 33
slide-34
SLIDE 34

Part II : Entropy removal

slide-35
SLIDE 35

Band insulator Mott insulator Spin disordered Hole doped Particle doped

T 100nK

Typical configurations of Lattice Fermions in a trap

n (r)

U  T  J

slide-36
SLIDE 36

n (r)

T  t  t 2 /U

~

t 2 /U  T

~ Esslinger, et.al, 2008 Science Bloch, et.al 2008

slide-37
SLIDE 37

ln2  0.69 ln3 1.1

n(r)

Density and entropy distribution

r /d r /d

s (r)

Band insulator Mott insulator (spin-1, ln3)

slide-38
SLIDE 38

Compression of trap Squeezing entropy to surfact Adiabaticity Temperature increase

slide-39
SLIDE 39

Isothermal compression => Entropy reduction

slide-40
SLIDE 40
slide-41
SLIDE 41

From Band insulator to Mott insulator: Turn down the lattice adiabatically 2 1

n(r)

Band insulator Mott insulator

n(r)

r /d

s(r)

Ln2

s(r)

r /d

40nK

S /N  0.016

T<<40nK

slide-42
SLIDE 42
  • 1. Can use a gapful phase of the system to push out

the entropy out from the bulk

  • 2. Remove the entropy by pushing it into a BEC

Or by evaporation.. But

  • 3. Always tighten the trap immediately to stop

entropy regeneration

Summary of our Cooling Strategy:

slide-43
SLIDE 43

Summary: Quantum Simulation Program => Lowest temperature regime ever achieved Quantum Many-body precision Measurement