bec meeting 23 may 2006bec.science.unitn.it/bec/4_activities/becmeeting2006/giuliano.pdf · bec...
TRANSCRIPT
-
BEC Meeting 23 May 2006
CNRINFM BEC CENTER & Physics Department
Cold Atoms in Optical LatticesG. Orso
-
●Sound propagation and oscillations of a superfluid Fermi gas in a 1D optical lattice, PRA 2005
●Formation of molecules near a Feshbach resonance in 1D optical lattice, PRL 2005●Twobody problem in periodic potentials, PRA 2006
●Umklapp collisions and oscillations of a trapped Fermi gas, PRL 2004
2. twobody problem in optical lattices
1. how to detect BCS transition in Fermi gases ?
-
●Sound propagation and oscillations of a superfluid Fermi gas in a 1D optical lattice, PRA 2005
●Formation of molecules near a Feshbach resonance in 1D optical lattice, PRL 2005●Twobody problem in periodic potentials, PRA 2006
●Umklapp collisions and oscillations of a trapped Fermi gas, PRL 2004
2. twobody problem in optical lattices
1. how to detect BCS transition in Fermi gases ?
Superfluid Fermi gas in a 1D optical lattice, PRL 2005
+
=
-
1D tight optical lattice + weak harmonic trapping
Kohn's theorem does not apply
collisional regime: oscillations killed by umklapp collisions
Relaxation rate from quantum Boltzmann equation
Question: what happens in interacting Fermi gases ?
1/uk
1uk
=C2 a2
d22 tℏ
F TF
,F4 t
T /F
F /4 t=5
21
PRL 2004
-
superfluid phase: persistent Josephson oscillations why ? phonons protected by energy gap
Can write hydrodynamic equations as for BEC. Ingredients: effective mass and compressibility (from BCS theory)
1ms=
∫ ∂2
∂ k z2
n0kd k
∫ n0 kd kn0k=2F−k
Quasimomentum distribution
eff. mass is density dependent. Recover BEC limit for
BEC: n0k=nk1/msb=1/m*
F≪4 t
●study of dynamical instability●application to trapped gasesPRA 2005
-
V ext z=s E R sin2 z /d no confinement
in xy directions
s laser intensity, ER=ℏ22
2 m d2recoil energy, d lattice spacing (0.11 m)
Bound states in 1D optical lattice
interparticle interactions modeled via swave pseudopotential Ur =gr ∂r rr
r=r1−r2 relative distance g=4ℏ2 a
mswave scattering
length
-
Green function for relative motion
1g= ∂∂r
r GE r r=0 GE r =−∑nnr n0
En−E
=CM Rr r V ext r1V ext r2=V CM RV r r
e.g. harmonic potential V ext r =12
m2 r2
COM separates
R=r1r2
2centerofmass coordinate
Binding energy given by algebraic equation
-
what about optical lattices ? e.g. 1D lattice
c.o.m. and relative motion coupledV opt z1V opt z2≠V CM Z V r z
!!
r , Z =∫ dZ ' GE r , Z ; 0, Z ' g ∂∂ r 'r 'r ' , Z ' r '=0
Solution via 2particle Green's function
Y Z =g∫dZ ' K E Z , Z ' Y Z '
regular kernel
Binding energy given by integral equation in COM variable
-
binding energy depends on quasimomentum
quasimomentum Q is conserved
invariant under Z Zd
f QZd =ei Q d f QZ solutions are Bloch states
K E Z , Z '
COM and relative motion coupled
For given Q, integral equation has been solved ●numerically PRL 2005●analytically (tightbinding) PRA 2006
-
s=20
10
5
0
Binding energy versus inverse scattering length
dots: lattice as perturbation
harmonic trap
d /acr binding energy vanishes
(Q=0)
-
Q=qB
Q=0
Critical value of scattering length
-
anisotropic 3D regime4t
∣Eb∣≪4 t
qz
qz
Regimes in tight 1D optical lattices
3D regime●bound state spread over many lattice sites●binding energy takes universal form ∣E b∣~
ℏ2
m C 1a− 1acr 2
renormalization of coupling constantg gC
*
-
g≫∣Eb∣≫4 t quasi2D regime
anisotropic 3D regime
dimensional crossover
4t∣Eb∣≪4 t
qz
qz
Regimes in tight 1D optical lattices
3D regime●bound state spread over many lattice sites●binding energy takes universal form ∣E b∣~
ℏ2
m C 1a− 1acr 2
renormalization of coupling constantg gC
*
-
quasi2D regime
∣Eb∣=ℏ0 exp−2 ∣a∣●tunneling is irrelevant: effective harmonic oscillator ●binding energy approaches asymptotic value
0.29
0
harmonic oscillator length
dacr=−C ln ℏ02 t
yields formula for
ℏ/m0 1/2
acr numericst. b.
-
BCS transition in 1D optical lattice
QUESTION: how is Tc modified by 1D lattice ?
1geff
=P∫ d3 q
231q
1exp q /T c
01
q=ℏ2 q p
2
2 mqz− dispersion relation with =F
effective coupling constant for Cooper pairs related to 2body scattering amplitude
geff
1geff
=m
4ℏ2ℜ[ 1f scE=2 ]
-
incident 2particle state
Scattering amplitude
f scE =a∫ dZE 0, Z ∂r rr , Z r=0
E r , Z =ei q p⋅r pqz z1−qzz2
E=ℏ2 q p
2
m2qz energy
Bloch statesqzz
Y Z ≡∂r rr , Z r=0
Y Z =E 0, Z g∫dZ ' K E Z , Z ' Y Z '
-
Tightbinding solution
Y Z ~A∑ j w2Z− j d w z=
11/41/2
exp [ −z2
22]
f sc E =aC
1−a /acra /2E
describes dimensional crossoverE
=i arccos(1 E/4t) (E8t)
ansatz:
Wannier state
-
anisotropic 3D regime (E 8t)
f scE =C
1/a−1/acri C E m /ℏ*
shift of Feshbach resonance
1geff
=m
4ℏ2 C 1a− 1acr density independent e.g. dilute BEC in optical lattices ∣a∣≪∣acr∣ geff≃Cg
lattice enters through effective mass and coupling constant
-
4 t1
geff=
m4ℏ2 C 1a 12 ln ℏ2 411−4 t /2
density dependent
change in behaviour occurs exactly at =4 t
f scE =C
1a
1
2 [ ln ℏE i]=0.29 quasi2D regime (E 8t)
-
Results for transition temperature
F4 t
T c0=
2
e−F F /4 t F exp [ −2 qzF C 1∣a∣− 1∣acr∣]3D regime T c0=0.61F exp [ −2 qzF C 1∣a∣− 1∣acr∣]F≪4 t
F4 t
T c0=
2 11− 4 tF F 4 t exp [− 2 1∣a∣− 1∣acr∣]quasi 2D regime
F≫4 tT c
0=0.43F ℏexp−2 ∣a∣KosterlitzThouless
-
Gorkov's corrections
Gorkov's corrections reduce mean field result T c /T c01
1/e
Van Hove singularity
-
Quantum phase model
SuperfluidMott Insulator QPT
●zero temperature●discs are finite size and tunneling is weak
t
phase of order parameter
Question: what is critical tunneling rate for transition to Mott phase ?
N j=NN j
in each disc relevant variables arenumber of atoms
j
HYDRODYNAMIC APPROACH
HQP=∑ j E c /2N j2−E J cos j1− j
[N j , j ]=1
Ec charging energy E J Josephson energy
(N 1)
-
weakly interacting BEC BCS superfluid
E J=t2
FN
E c0.81 E J superfluidE c0.81 E J Mott insulator
tc~FN
Ec=2
N
E JB=t N
tcB~
N 2
BradleyDoniach, PRB (1984)
Complete analogy with BEC:
...but Josephson term is different !
Cooper pairs single atom
-
3body losses quenched by Fermi statistics
Question: can this QPT be achieved with cold gases ?Answer: only if critical tunneling rate large compared to atom loss rate due to 3body recombination
● BEC: NO, unless● BCS superfluids: YES, up to N~103−104
N~1−5
Ultracold gases
-
Conclusions
umklapp collisions in Fermi gasespropagation of sound in superfluid Fermi gases twobody problem BCS transition temperature SuperfluidMott insulator QPT in Fermi superfluids
-
Conclusions
umklapp collisions in Fermi gasespropagation of sound in superfluid Fermi gases twobody problem BCS transition temperature SuperfluidMott insulator QPT in Fermi superfluids