Y. Castin, LKB - ENS, Paris: Basics of Bose-Einstein Condensation Theory

Download as pdf or txt
Download as pdf or txt
You are on page 1of 18

BASICS OF BOSE-EINSTEIN CONDENSATION THEORY

Y. Castin, LKB - ENS, Paris

OUTLINE

atoms: waves and particles


the Bose law
when the Bose gas becomes degenerate
how to reach Bose-Einstein condensation
atomic interactions and Gross-Pitaevskii equation

1
ATOMS: WAVES AND PARTICLES
Analogy with optics:
Object optics atomic physics

field E(r, t), B(r, t) (r, t)


1 2 h2
 
equation of motion 2 t B = 0 iht =
c 2m
particle photon atom
1
energy h mv 2
2
momentum hk p = mv
2 h h
wavelength = = =
k hk p
hk2
dispersion relation = ck =
2m

2
Values in an ordinary gas:
equipartition of energy:
1 2 1
mhvxi = kB T
2 2
sodium atoms at 300 K:
vx = 300 m/s
= 5 1011 m
With Sisyphus cooling:
1 m.

3
ATOMIC MODES IN A BOX

Energy levels of an atom in a box:


periodic boundary conditions:
(x + L, y, z) = (x, y + L, z) = (x, y, z + L)
= (x, y, z).

quantisation of wavevectors:
(x, y, z) ei(kxx+ky y+kz z)
2
k = q
L
quantisation of energy:
h2  2 2 + q2

k = q + q
2mL2 x y z

4
THE BOSE LAW

Indistinguishable particles in quantum theory are:


bosons:
P |iB = |iB

or fermions:
P |iF = ()|iF .

Configuration defined by a set of occupation numbers {n}

Example: two spin 1/2 particles of opposite spin:


|iB |+i |i + |i |+i
|iF |+i |i |i |+i
|+i |i meaningless

5
Thermodynamics of the ideal Bose gas:

1 P ()n
Proba({n}) = e

where = 1/(kB T ) and is the chemical potential.

Bose law for the occupation number:


1
hni =
e() 1
so that
< < 0.

6
Lower limit for is non-degenerate regime:
3 h 2k2/2m
hn~k i ' e

in a large box, where


s
2h2
=
mkB T
is the thermal de Broglie wavelength.
The coherence length of the gas is .

7
WHEN THE BOSE GAS BECOMES DEGENERATE
3  1

Saturation of excited state population:


0
X X 1 0
N hni < Nmax
e (0) 1
6=0 6=0

For a large cubic box


h2
L  , i.e. kB T  ,
2mL2
0
X 1 L3
Nmax = ' 2.612 3
e h2k2/2m 1
~
k6=~0

8
0
If N > Nmax ...
0
. . . there are at least N Nmax atoms in the ground mode
of the box.
A condensate forms if:

3 > 2.612 . . . Einstein, 1925


Totally counter-intuitive for Boltzmann statistics.
In a harmonic potential:
3
kB T

0
Nmax ' 1.202
h
where is the geometric mean of the trap frequencies.

9
0
Even in a trap one has for N = Nmax :

(~0 )3 ' 2.612

Below Tc: condensate fraction

0 3/2
N0 N Nmax T

' ' 1 box
N N Tc
 3
T
' 1 harmonic trap
Tc
Realistic examples:

T /Tc = 1/2 everyday

T /Tc = 1/4 the good days

10
BEC in position space: kB T = 20h N = 500 to 32000
4
10

2
10
(r) [a0 ]
3

0
10

2
10

4
10
0 10 20
r [a0]

11
Results of JILA:

0.8

0.6
N0/N

0.4

0.2

0
0 0.2 0.4 0.6 0.8 1
0
T/T c

12
HOW TO REACH BOSE-EINSTEIN CONDENSATION

The problem of solidification:


For air with pressure 1 atm:
Tc ' 0.4K
but then one expects a solid phase.

He4 does not solidify. Experiences superfluid transition


at 2K
but is a liquid, not a gas (condensate fraction < 0.1).

Only polarized hydrogen is gaseous at 1 atm, 0 K.

13
Low density route: use of metastability
2-body elastic collisions ensure thermalisation:
elas

3-body collisions form molecules:


inel 2
but are much slower at low density!

the obtained condensate is metastable.

Price to pay: ultralow temperatures


Tc 40nK to 1K.

14
How to cool ?
laser cooling alone not yet succeeded:
2
opt =
kL
and bad effects of light when 3opt 1.

forced evaporative cooling: remove atoms in high energy


tails, let gas rethermalize, and so on

efficient if elas
> 100.
loss

15
EFFECT OF ATOMIC INTERACTIONS

Not ideal
Bose gas!

Coherence
length!!

16
How to characterize the interaction potential?
by its r dependence: by its scattering length:
h2
500

250 (r) = 0
m
0
V(r12) /kB [K]

250
(r) = C0 + C1/r
500
a
1
750
0 0.5
r12 [nm]
1 1.5
r

Typical values
a = 50 nm (87Rb) a = 1.5 nm (7Li)
but a can be tuned.

17
THE GROSS-PITAEVSKII EQUATION

" #
h2
iht(~
r , t) = + U (~ r , t)|2 (~
r ) + gN0|(~ r , t)
2m

Comes from mean field for model interaction potential

V (~
r ) = g(~
r )r(r )

4h2
with coupling constant g = a.
m
Explains almost everything, including superfluidity.

18

You might also like