The Hubbard Model for Dummies

Introduction
Some math
2
nd
quantization
The partition function
Limiting cases
t = 0
U = 0
Mean field theory
Matrix notation and the t-J model
Extensions of the Hubbard model
Bruce Patterson
13.6.07
The Hubbard Model: Introduction
J. Hubbard, Proc. Roy. Soc. London, A266, 238 (1963).
Simplest model of interacting particles in a lattice: extension of the tight-
binding model to include short-range el-el interactions.
One-band Hubbard Hamiltonian (one orbital per site):
Half-filling: on average one electron per site.
Electrons may hop from site to site, but for large U/t, we get a Mott insulator.
Note: One of the first correlated electron calculations: Heitler+London treatment
of H
2
-molecule (1927).
!
H = "t c
j#

ij #
$
c
i#
+U n
i%
i
$
n
i&
<ij> implies
adjacent sites
hopping term interaction term
Second Quantization
Boson annihilation and creation operators for the harmonic oscillator:
From , we verify that and that the oscillator Hamiltonian is:
From the ground state , we build up the excited states:

!
a "
m#
2h
x + i
1
2m#h
p
a

"
m#
2h
x $i
1
2m#h
p

!
x, p
[ ]
= ih
!
a, a

[ ]
=1

!
H
osc
=
1
2m
p
2
+
m"
2
2
x
2
= h" a

a +
1
2
#
$
%
&
'
( = h" n +
1
2
#
$
%
&
'
(
n=a

a = number operator.
!
0

!
a

n = n +1 n +1 H
osc
n = h" n +
1
2
#
$
%
&
'
( n
Electron states for the (N=8) Hubbard model in terms of fermionic operators:
!
c
1"

0 = ",#,#,#,#,#,#,#
c
5$

0 = #,#,#,#,$,#,#,#
c
5$

c
1"

0 = ",#,#,#,$,#,#,#
c
4"

c
4$

0 = #,#,#,"$,#,#,#,#
!
c
1"
0 = 0
c
1"

c
1"

0 = 0
c
i#
,c
j# '
{ }
$ c
i#
c
j# '
+ c
j# '
c
i#
= 0
c
i#

,c
j# '

{ }
= 0
c
i#
,c
j# '

{ }
=%
ij
%
## '
{} = anticommutator
!
c
1"

c
1"
",#,#,#,#,#,#,# $ n
1"
",#,#,#,#,#,#,# =1",#,#,#,#,#,#,#
c
5%

c
5%
",#,#,#,#,#,#,# $ n
5%
",#,#,#,#,#,#,# = 0",#,#,#,#,#,#,#
Partition Function
Grand Canonical Ensemble: particles can be exchanged with a resevoir.
Thermal expectation value of operator A:
Consider a single state with 0 or 1 electrons with energy !:
Average occupation:
!
partition function: Z " Tr e
#$ H#n ( )
[ ]
$ "
1
kT
" "chemical potential"
!
H ="c

c ="n states : 0 , 1
{ }
!
Z = 0 e
"# H"n ( )
0 + 1e
"# H"n ( )
1 =1+ e
" # " ( )
!
A =
1
Z
Tr Ae
"# H"n ( )
[ ]
!
n =
1
Z
0 ne
"# H"n ( )
0 + 1 ne
"# H"n ( )
1
[ ]
=
1
Z
0 + e
"# $ " ( )
[ ]
=
e
"# $ " ( )
1+ e
"# $ " ( )
=
1
e
# $ " ( )
+1
= Fermi - Dirac distribution
Hubbard Model with t=0
No hopping implies independent sites. We thus consider a single site and
calculate its average occupation.
!
possible states = 0 , " , # , "#
{ }
!
H =Un
"
n
#
!
Z = "
"
#
e
$% H$n ( )
" =1+ e
%
+ e
%
+ e
$%U+2%
n =
1
Z
"
"
#
n
&
+ n
'
( )
e
$% H$n ( )
" =
1
Z
0 + e
%
+ e
%
+ 2e
$%U+2%
[ ]
=
2 e
%
+ e
$%U+2%
( )
1+ 2e
%
+ e
$%U+2%
Note: Since , at half-filling, the energy to add another particle to the
site jumps by U. This is the Hubbard gap.
0,6
0,8
1
1,2
1,4
1,6
1,8
2
0 0,5 1 1,5 2 2,5
T=0.05
T=0.1
T=0.5
<
n
>

U=1
"half filling"
"Hubbard gap"
(=U)
!
=
"E
"n
Hubbard Model with U=0
Non-interacting limit: we go into reciprocal space:
where k takes the values k
n
=2"n/N. (1-d, periodic boundary conditions)
!
c
k"

=
1
N
e
ik# l
l
$
c
l"

!
H = "t c
j#

c
l#
jl #
$
=
"t
N
e
ikj
e
"ik' l
c
k#

c
k'#
jl #
$
k,k'
$
=
"t
N
c
k#

c
k'#
e
ikj
e
"ik' j +1 ( )
+ e
ikj
e
"ik' j"1 ( )
[ ]
= "t c
k#

c
k'#
e
"ik'
+ e
ik'
[ ]
1
N
e
i(k"k' ) j
j
$
k,k',#
$
j
$
k,k',#
$
= "t c
k#

c
k'#
2cosk'
[ ]
%
k,k'
k,k',#
$
= &
k
n
k#
k#
$
&
k
= "2t cosk
For U=0, the energy-eigenvalues of H are:
The bandwidth is equal to 4t.
!
"
k
= #2t cosk k
n
=
2$n
N
n = 0,1,2,...N #1
-3
-2
-1
0
1
2
3
-5 0 5
!
k



[
t
]
k [2"/N]
4t
The Mott Transition:
U>>t: diagonalize H in real space - at half-filling we obtain a static
(insulating) system.
t>>U: diagonalize H in reciprocal space - obtain itinerant electron band,
width 4t.
At some intermediate value of U/t, there will be a metal-to-insulaltor
transition: the Mott transition.
Mean Field Theory
We wish to treat the full Hubbard model (in 1-d):
Problem: the second term is proportional to c
4
.
Mean field approximation: each electron feels an average interaction from all others.
This ignores possibly important electron correlations.
Write:
and assume # is small, and analogously for the down spins. Expand:
!
H = "t c
j#

c +U n
l$
n
l%
l
&
jl #
&
!
n
l"
= n
"
+ n
l"
# n
" [ ]
$ n
"
+%
l"
!
n
l"
n
l#
$ n
"
n
#
+%
l"
n
#
+%
l#
n
"
$ n
l"
n
#
+ n
l#
n
"
& n
"
n
#
In reciprocal space:
H is now diagonal:
We define spin-dependent Fermi wavevectors:
After integrating, we obtain for the total energy:
!
c
l"

=
1
N
e
#ikl
k
$
c
k"

!
H = "2t cosk n
k#
+ n
k$
( )
+U n
k#
n
$
+ n
k$
n
# ( ) [ ]
"UN n
#
k
%
n
$
= &
k#
n
k#
+&
k$
n
k$
[ ]
"UN n
#
k
%
n
$
&
k#
' "2t cosk +U n
$
!
=
k"
#
N
$
dk
0
k
F"
%
k
F"
=
$ n
"
N
!
E
tot
=
"2Nt
#
sin# n
$
+ sin# n
% [ ]
+UN n
$
n
%
For half-filling: write E
tot
as a function of
!
n
"
+ n
#
=1
!
"E
tot
n
# ( )
4Nt
= $sin" n
#
+
U"
4t
n
#
1$ n
# ( )
-0,3
-0,2
-0,1
0
0,1
0,2
0,3
0 0,2 0,4 0,6 0,8 1
E
!
/
4
N
t
<n
up
>
U!/4t = 3
3.25
3.5
3.75
4
4.25
4.5
4.75
5
small U $ a paramagnet
large U $ a ferromagnet
This is completely wrong!
The 1-dimensional Hubbard model has
been exactly solved by Lieb and Wu, PRL
20, 1445 (1998). At infinitesimal U/t an
antiferromagnetic state forms.
!
n
"
Hubbard Model in Matrix Notation
Toy Model
(simplified H
2
molecule)
Two orthonormal orbitals separated by some distance:
Add a single electron, which moves with an amplitude -t.
$ Tight-binding" Hamiltonian:
Eigenvectors:
Energy eigenvalues: Note: %
+
is the ground state.
!
H =
0 "t
"t 0
#
$
%
&
'
(
!
"

=
1
2
"
1
"
2
( )
=
1
2
1
1
#
$
%
&
'
(

!
"

= mt
!
"
1
=
1
0
#
$
%
&
'
(
, "
2
=
0
1
#
$
%
&
'
(
!
"
1
= #,$ , "
2
= $,# schematically:
We now consider two electrons:
Hamiltonian
ionic states
covalent states
!
",# = c
2#

c
1"

0
#," = c
2"

c
1#

0
"#,$ = c
1#

c
1"

0
$,"# = c
2#

c
2"

0
!
H =
0 0 "t "t
0 0 +t +t
"t +t U 0
"t +t 0 U
#
$
%
%
%
%
&
'
(
(
(
(
),*
*,)
)*,+
+,)*
Note: +t due to fermion
transposition
Diagonalize H:
Energy levels:
!
"

=
U
2

U
2
+16t
2
2
#

=
$,% & %,$ &
"

2t
$%,' + ' $%
[ ]
(
)
*
+
,
-
2 + "

2
/ 2t
2
( )
"
cov
= 0 #
cov
=
1
2
$,% + %,$
( )
"
ion
=U #
ion
=
1
2
$%,' & ',$%
( )
-2
0
2
4
6
8
0 1 2 3 4 5 6 7 8
!
/
t
U/t
"
+
"
ion
"
-
"
cov
Consider the limit of large U:
&
-
is now largely covalent, with a small ionic admixture.
With downfolding, we will project out the interesting, low-
energy, covalent-like part of Hilbert space.
!
"
cov
= 0
"
#
$
#4t
2
U
A note on matrix inversion by partitioning:
Ref: Ch. 2.7, Numerical Recipes, 2nd Ed. (1992)
!
A =
A0 A1
A2 A3
"
#
$
%
&
'
We divide a matrix A into the
blocks A0-A3.
!
A
"1
=
B0 B1
B2 B3
#
$
%
&
'
(
The inverse can also be written in
block form:
!
B0 = A0 " A1# A3
"1
# A2
( )
"1
B1= "B0# A1# A3
"1
B2 = "A3
"1
# A2# B0
B3 = A3
"1
" B2# A1# A3
"1
We begin downfolding by partitioning H into blocks:
The Greens function can also be partitioned:
and the (energy-dependent) covalent part G
00
can be written:
This looks like the Greens function for the effective Hamiltonian:
!
H =
0 0 "t "t
0 0 +t +t
"t +t U 0
"t +t 0 U
#
$
%
%
%
%
&
'
(
(
(
(
)
H
00
T
01
T
10
H
11
#
$
%
&
'
(
H
00
= covalent Hamiltonian
H
11
= ionic Hamiltonian
T
ij
=covalent-ionic transitions
!
G "
( )
# " $ H
( )
$1
=
" $ H
00
$T
01
$T
10
" $ H
11
%
&
'
(
)
*
$1
#
G
00
G
01
G
10
G
11
%
&
'
(
)
*
!
H
eff
"
( )
= H
00
+ T
01
" # H
11
( )
#1
T
10
$ H
eff
"
0
( ) !
0
is a typical covalent energy.
!
G
00
"
( )
= " # H
00
+ T
01
" # H
11
( )
#1
T
10 [ ] ( )
#1
We evaluate the effective Hamiltonian between the covalent states:
which diagonalizes to:
These states correspond to singlet and triplet states of the H
2
molecule.
The other triplet states, , also have zero energy. (There is no
Coulomb energy, and the electrons cannot hop, due to the Pauli principle.)
!
H
eff
"
0
( )
= H
00
+ T
01
"
0
# H
11
( )
#1
T
10
=
0 0
0 0
$
%
&
'
(
)
+
#t #t
t t
$
%
&
'
(
)
"
0
#U 0
0 "
0
#U
$
%
&
'
(
)
#1
#t t
#t t
$
%
&
'
(
)
H
eff
0
( )
= #
2t
2
U
1 #1
#1 1
$
%
&
'
(
)
*,+
+,*
!
"
s
= #
4t
2
U
$
s
%
1
2
&,' # ',&
( )
"
t
= 0 $
t
=
1
2
&,' + ',&
( )
Note that we choose !
0
=0.
!
"," , #,#
Recall that &
s
includes
a small ionic admixture.
Kinetic Exchange
Anti-parallel alignment of the spins is favored, since it allows the
electrons to hop to the neighboring site.
We can write H
eff
in terms of the fermion operators:
Consider the properties of the Pauli spin matrices:
!
H
eff
= "
2t
2
U
c
1#

c
1#
c
2$

c
2$
"c
1$

c
1#
c
2#

c
2$
"c
1#

c
1$
c
2$

c
2#
+ c
1$

c
1$
c
2#

c
2#
( )

!
" # + #
1
0
$
%
&
'
(
)
* # + #
0
1
$
%
&
'
(
)
r
S =
1
2
,
x
)
x + ,
y
)
y + ,
z
)
z
( )
,
x
=
0 1
1 0
$
%
&
'
(
)
,
y
=
0 +i
i 0
$
%
&
'
(
)
,
z
=
1 0
0 +1
$
%
&
'
(
)
,
x
= c
"

c
*
+ c
*

c
"
( )
= m
,
y
= +i c
"

c
*
+c
*

c
"
( )
= i m
,
z
= c
"

c
"
+c
*

c
*
( )
= n
"
+ n
*
( )
=
We can therefore rewrite H
eff
:
The Hubbard Hamiltonian, for large U/t, has led us to an
antiferromagnetic exchange coupling J=4t
2
/U.
This is the basis of the t-J model.

!
H
eff
=
4t
2
U
r
S
1
"
r
S
2
#
n
1
n
2
4
$
%
&
'
(
)
* Heisenberg Hamiltonian
!
n
i
" n
i#
+ n
i$
We have solved the 4x4 Hubbard Hamiltonian by exact diagonalization. For
larger systems, matrix expressions for H grow quickly in size. For N sites:
For 30 sites and 15 spins of each orientation, this dimension is
2.4 x 10
16
.
This is why we use 2
nd
quantization.
No exact solution the Hubble model in >1 dimension is known.
!
dimension of Hilbert space =
N
n
"
#
$
%
&
'
(
)
N
n
*
#
$
%
&
'
(
!
N
n
"
#
$
%
&
'
(
N!
n! N ) n
( )
!
2
1
"
#
$
%
&
'
*
2
1
"
#
$
%
&
'
= 4
(mutually inconsistent) approximate solutions of the 2-D
Hubbard Model
Results of 6 groups,
compiled by M.P. March,
Condensed Matter
Physics, Wiley, 2000.
!
" # n $1
" = 0 %half - filling
Extensions of the Hubbard Model
Three-Band Hubbard Model
To describe the 2-d CuO
2
layers in superconducting cuprates, this model
includes:
Cu(3dx
2
-y
2
) states (U
d
)
O(2px,y) states (U
p
)
their nn interaction (U
pd
)
For large U
pd
, this Hamiltonian reduces to the t-J model.
2-d hole-pairing:
AF-coupling causes holes to form pairs.
unpaired: 8 bonds broken paired: 7 bonds broken
Charge-Spin separation:
In 1-d, Fermi liquid theory breaks down, due to the destruction of
the Fermi surface (Fermi points) by an arbitrarily small perturbation.
In this case, separate gapless charge (holon) and spin (spinon)
excitations arise. This state is a Luttinger liquid.
Creation in photoemission
of a holon and a spinon.
The Hubbard model with an infinite number of dimensions
Each site of the Hubbard model can then be treated as an isolated
impurity interacting with an electron resevoir. This problem can be
attacked with the dynamical mean-field theory.
Ref: G. Kotliar and D. Vollhardt, Phys. Today March, 53 (2004).
At intermediate U/W (=U/t), we now see a
sharp DOS peak at the Fermi level (Kondo
resonance).
The Hubbard-Holstein model (with electron-phonon interaction)
Ref: W. Koller, et al, cond-mat/0312367v2 (2004).
!
H
H" H
= "t c
j#

c
i#
+U n
i$
i
%
ij #
%
n
i&
" g b
i

+ b
i
( )
n
i#
i#
%
+'
0
b
i

i
%
b
i
!
b
i

= creation operator of the

local phonon mode at site i
!
g = electron - phonon
coupling constant
In the limit of large '
0
, the
H-H model maps onto the
Hubbard model with
H
eff
=U-2g
2
/ '
0
.
Introductory References to the Hubbard Model
E. Koch, Electronic Structure of Matter: Electron Correlations, in
Probing the Nanoworld, lecture manuscripts of the 38th Spring
School, Jlich (2006).
R.T. Scalettar, Elementary Introduction to the Hubbard Model,
lecture notes, UC Davis,
http://leopard.physics.ucdavis.edu/rts/p210/hubbard7.pdf
M.P. March, Condensed Matter Physics, Wiley, 2000.
Some examples of 2-electron calculations:
!
c
1"

c
2"
",# = c
1"

c
2"
c
2#

c
1"

0 = $c
1"

c
2#

c
2"
c
1"

0 = +c
1"

c
2#

c
1"

c
2"
0 = 0
c
1#

c
2#
%,"# = c
1#

c
2#
c
2#

c
2"

0 = c
1#

c
2"

0 $c
1#

c
2#

c
2#
c
2"

0 = $c
2"

c
1#

0 = $#,"
c
2"

c
1"
",# = c
2"

c
1"
c
2#

c
1"

0 = $c
2"

c
2#

c
1"
c
1"

0 = $c
2"

c
2#

0 + c
2"

c
2#

c
1"

c
1"
0 = c
2#

c
2"

0 = %,"#