a

r
X
i
v
:
c
o
n
d
-
m
a
t
/
0
2
1
0
5
1
9
v
2


[
c
o
n
d
-
m
a
t
.
m
e
s
-
h
a
l
l
]


3

J
a
n

2
0
0
6
Non Equilibrium Greens Functions for Dummies:
Introduction to the One Particle NEGF equations
Magnus Paulsson

Dept. of micro- and nano-technology, NanoDTU,

Technical University of Denmark
February 1, 2008
Abstract
Non equilibrium Greens function methods are regularly used to calculate current and charge
densities in nanoscale (both molecular and semiconductor) conductors under bias. This method is
mainly used for ballistic conduction but may be extended to include inelastic scattering. In this tutorial
paper the NEGF equations for the current and charge density matrix are derived and explained in a
hopefully clear way.
1 Introduction
Non equilibrium Greens function methods are regularly used to calculate current and charge densities
in nanoscale (both molecular and semiconductor) conductors under bias. An overview of the theory of
molecular electronics can be found in Ref. [1] and for semiconductor nanoscale devices see Ref. [2].
The aim of this text is to provide some intuitive explanations of one particle Greens functions in a
compact form together with derivations of the expressions for the current and the density matrix. It is
not intended as a complete stand-alone tutorial, but rather as a complement to Ref. [1, 2, 3, 4].
2 Greens functions
Discrete Schrodinger equation:
H|n = E|n (1)
We divide the Hamiltonian and wavefunction of the system into contact (H
1,2
, |
1,2
) and device (H
d
,
|
d
) subspaces:
_
_
H
1

1
0

1
H
d

2
0
2
H
2
_
_
_
_
|
1

|
d

|
2

_
_
= E
_
_
|
1

|
d

|
2

_
_
(2)

mpn@mic.dtu.dk
1
where
1,2
describes the interaction between device and contacts. In general we have N contacts (H
1,,N
)
connecting (
1,,N
) the device H
d
to the reservoirs. Here we will assume that the contacts are independent,
i.e., there are no cross terms () between the dierent contacts.
We dene the Greens function
1
:
(E H) G(E) = I (3)
2.1 Why do we want to calculate the Greens function?
The Greens function gives the response of a system to a constant perturbation |v in the Schrodinger
equation:
H| = E| + |v (4)
The response to this perturbation is:
(E H) | = |v (5)
| = G(E)|v (6)
Why do we need the response to this type of perturbation? Well, it turns out that its usually easier
(see next section) to calculate the Greens function than solve the whole eigenvalue problem
2
and most
(all for the one-particle system) properties of the system can be calculated from the Greens function.
E.g., the wavefunction of the contact (|
2
) can be calculated if we know the wavefunction on the device
(|
d
). From third row of Eq. 2:
H
2
|
2
+
2
|
d
= E|
2
(7)
(E H
2
) |
2
=
2
|
d
(8)
|
2
= g
2
(E)
2
|
d
(9)
where g
2
is the Greens function of the isolated contact 2 ((E H
2
)g
2
= I).
It is important to note that since we have an innite system, we obtain two types of solutions for the
Greens functions
3
, the retarded and the advanced
4
solutions corresponding to outgoing and incoming
waves in the contacts.
Notation: We will denote the retarded Greens function with G and the advanced with G

(and maybe
G
R
and G
A
occasionally). Here, CAPITAL G denotes the full Greens function and its sub-matrices G
1
,
G
d
, G
1d
etc. Lowercase is used for the Greens functions of the isolated subsystems, e.g., (E H
2
) g
2
= I.
Note that by using the retarded Greens function of the isolated contact (g
2
) in Eq. 9 we obtain the
solution corresponding to a outgoing wave in the contact. Using the advanced Greens function (g

2
) would
give the solution corresponding to an incoming wave.
1
Others may (and do) use the opposite sign.
2
Especially for innite systems.
3
When the energy coincides with energy band of the contacts there are two solutions corresponding to outgoing or
incoming waves in the contacts.
4
In practice these two solutions are usually obtained by adding an imaginary part to the energy. By taking the limit to
zero of the imaginary part one of the two solutions is obtained. If the limit 0
+
is taken the retarded solution is found,
0

gives the advanced. This can be seen from the Fourier transform of the time dependent Greens function.
2
2.2 Self-Energy
The reason for calculating the Greens function is that it is easier that solving the Schrodinger equation.
Also, the Greens function of the device (G
d
) can be calculated separately without calculating the whole
Greens function (G). From the denition of the Greens function we obtain:
_
_
E H
1

1
0

1
E H
d

2
0
2
E H
2
_
_
_
_
G
1
G
1d
G
12
G
d1
G
d
G
d2
G
21
G
2d
G
2
_
_
=
_
_
I 0 0
0 I 0
0 0 I
_
_
(10)
Selecting the three equations in the second column:
(E H
1
) G
1d

1
G
d
= 0 (11)

1
G
1d
+ (E H
d
) G
d

2
G
2d
= I (12)
(E H
2
) G
2d

2
G
d
= 0 (13)
We can solve Eqs. 11 and 13 for G
1d
and G
2d
:
G
1d
= g
1

1
G
d
(14)
G
2d
= g
2

2
G
d
(15)
substitution into Eq. 12 gives:

1
g
1

1
G
d
+ (E H
d
) G
d

2
g
2

2
G
d
= I (16)
from which G
d
is simple to nd:
G
d
= (E H
d

1

2
)
1
(17)
where
1
=

1
g
1

1
and
2
=

2
g
2

2
are the so called self-energies.
Loosely one can say that the eect of the contacts on the device is to add the self-energies to the
device Hamiltonian since when we calculate the Greens function on the device we just calculate the
Greens function for the eective Hamiltonian H
effective
= H
d
+
1
+
2
. However, we should keep in mind
that we can only do this when we calculate the Greens function. The eigen-values and -vectors of this
eective Hamiltonian are not quantities we can interpret easily.
For normal contacts, the surface Greens functions g
1
and g
2
used to calculate the self-energies are
usually calculated using the periodicity of the contacts, this method is described in detail in appendix B
of Ref. [3] and in section 3 of Ref. [2].
2.3 The spectral function
Another important use of the Greens function is the spectral function:
A = i
_
G G

_
(18)
which gives the DOS and all solutions to the Schrodinger equation!
To see this we rst note that for any perturbation |v we get two solutions (|
R
and |
A
) to the
perturbed Schrodinger equation:
(E H)| = |v (19)
3
from the advanced and retarded Greens functions:
|
R
= G|v (20)
|
A
= G

|v (21)
The dierence of these solutions (|
R
|
A
= (G G

)|v = iA|v) is a solution to the Schrodinger

equation:
(E H)(|
R
|
A
) = (E H)(G G

)|v = (I I)|v = 0 (22)

which means that | = A|v is a solution to the Schrodinger equation for any vector |v!
To show that the spectral function actually gives all solutions to the Schrodinger equation is a little
bit more complicated and we need the expansion of the Greens function in the eigenbasis:
G =
1
E + i H
=

k
|kk|
E + i
k
(23)
where the is the small imaginary part (see footnote 3), |ks are all eigenvectors
5
to H with the corre-
sponding eigenvalues
k
. Expanding the spectral function in the eigenbasis gives:
A = i
_
1
E + i H

1
E i H
_
(24)
= i

k
|kk|
_
1
E + i
k

1
E i
k
_
(25)
=

k
|kk|
2
(E
k
)
2
+
2
(26)
where is our innitesimal imaginary part of the energy. Letting go to zero gives:
A = 2

k
(E
k
) |kk| (27)
(here (E
k
) is the delta function) which can be seen since
2
(E
k
)
2
+
2
goes to zero everywhere but at
E =
k
, integrating over E (with a test function) gives the 2 (E
k
) factor. Eq. 27 shows that the
spectral function gives us all solutions to the Schrodinger equation.
3 Response to an incoming wave
In the non-equilibrium case, reservoirs with dierent chemical potentials will inject electrons and occupy
the states corresponding to incoming waves in the contacts. Therefore, we want to nd the solutions
corresponding to these incoming waves.
Consider contact 1 isolated from the other contacts and the device. At a certain energy we have
solutions corresponding to an incoming wave that is totally reected at the end of the contact. We will
denote these solutions with |
1,n
where 1 is the contact number and n is a quantum number (we may
5
Normalized!
4
have several modes in the contacts). We can nd all these solutions from the spectral function a
1
of the
isolated contact (as described above).
Connecting the contacts to the device we can calculate the wavefunction on the whole system caused by
the incoming wave in contact 1. To do this we note that a wavefunction should be of the form |
1,n
+|
R

where |
1,n
is the totally reected wave and |
R
is the retarded response of the whole system. Putting
in the anzats |
1,n
+ |
R
into the Schrodinger equation gives:
_
_
H
1
+
1
+
H
d
+

1
+

2
+
H
2
+
2
_
_
_
|
1,n
+ |
R

_
= E
_
|
1,n
+ |
R

_
(28)
_
_
E|
1,n
+

1
|
1,n
+
0
_
_
+
_
_
H
1
+
1
+
H
d
+

1
+

2
+
H
2
+
2
_
_
|
R
= E
_
|
1,n
+ |
R

_
(29)
H|
R
= E|
R

1
|
1,n
(30)
(note the slight change in notation) and we see that |
R
is noting else the response of the whole system
to a perturbation of

1
|
1,n
, c.f., Eq. 9:
|
R
= G

1
|
1,n
(31)
It is important to realize that the scattering states generated from Eq. 31, using all possible incoming
waves from each contact, form a complete
6
ON set of solutions to the full Schrodinger equation[5]. Note
that we have chosen the retarded response which means that the only part of the wave that is traveling
towards the device is the incoming wave (part of |
1,n
). We will make full use of this fact below.
It will be useful to have the expressions for the device wavefunction |
d
and contact wavefunction
(|
1,2
). The device part is straightforward:
|
d
= G
d

1
|
1,n
(32)
and from Eq. 9 or Eq. 15:
|
2
= g
2

2
|
d
= g
2

2
G
d

1
|
1,n
(33)
Note, to calculate the wavefunction in the contact containing the incoming wave (contact 1) we need to
add the incoming wave, giving a slightly more complicated expression:
|
1
=
_
1 + g
1

1
G
d

1
_
|
1,n
(34)
Knowing the wavefunctions corresponding to incoming waves in dierent contacts enables us to ll up
the dierent solutions according to the electron reservoirs lling the contacts.
4 Charge density matrix
In the non equilibrium case we are often interested in two quantities: the current and the charge density
matrix. Lets start with the charge density (which allows us to use a self-consistent scheme to describe
charging).
6
Except for localized states in the device region.
5
The charge density matrix is dened as:
=

k
f(k, )|
k

k
| (35)
where the sum runs over all states with the occupation number f(E
k
, ) (pure density matrix) (note the
similarity with the spectral function A, in equilibrium you nd the density matrix from A and not as
described below). In our case, the occupation number is determined by the reservoirs lling the incoming
waves in the contacts such that:
f(E
k
,
1
) =
1
1 + e
(E
k
1)/kBT
(36)
is the Fermi-Dirac function with the chemical potential (
1
) and temperature (T) of the reservoir respon-
sible for injecting the electrons into the contacts.
The wavefunction on the device given by an incoming wave in contact 1 (see Eq. 32) is:
|
d,k
= G
d

1
|
1,k
(37)
Adding up all states from contact 1 gives:

d
[contact 1] =

_
E=
dE

k
f(E,
1
)(E E
k
)|
d,k

d,k
| (38)
=

_
E=
dE f(E,
1
)

k
(E E
k
)G
d

1
|
1,k

1,k
|
1
G

d
(39)
=

_
E=
dE f(E,
1
)G
d

1
_

k
(E E
k
)|
1,k

1,k
|
_

1
G

d
(40)
= [Eq. 27] =

_
E=
dE f(E,
1
)G
d

1
a
1
2

1
G

d
(41)
introducing the new quantity
1
=

1
a
1

1
= i
_

1
_
we obtain the simple formula:
[from contact 1] =
1
2

_
E=
dE f(E,
1
)G
d

1
G

d
(42)
The total charge density thus becomes a sum over all contacts:
=
2 (for spin)
2

_
E=
dE

i
f(E,
i
)G
d

i
G

d
(43)
6
5 Probability Current
Having dierent chemical potentials in the reservoirs lling the contacts gives rise to a current. In the
next section we will calculate this current in a similar way as the charge density was calculated. But to
do this we need an expression for the current from the wavefunction.
In the continuum case we can calculate the current from the velocity operator. However, for a discrete
Hamiltonian it is not so clear what the velocity operator is. Therefore, we derive an expression for the
current from the continuity equation (using two contacts). In steady-state, the probability to nd an
electron on the device (

i
|
i
|
2
where the sum runs over the device subspace) is conserved:
0 =

i
|
i
|
2
t
=

i
|ii|
t
=

i
_
|i
t
i| + |i
i|
t
_
(44)
=
i
h

i
(|H|ii| |ii|H|) =
i
h
(|H|
d

d
|H|) (45)
=
i
h
_
|H
d
+
1
+
2
|
d

d
|H
d
+

1
+

2
|
_
(46)
=
i
h
__

1
|
1
|
d

d
|

1
|
1

_
+
_

2
|
2
|
d

d
|

2
|
2

__
(47)
We interpret the term in the rst (square) bracket as the incoming probability current into the device
from contact 1 and the second bracket from contact 2. Generalizing to an arbitrary contact j gives us the
electric current (at one energy) as the charge (-e) times the probability current:
i
j
=
ie
h
_

j
|
j
|
d

d
|

j
|
j

_
(48)
where i
j
is dened as positive for a current from the contacts into the device. We can now put in the
expressions for the wavefunctions in the same way as for the density matrix.
6 Electrical Current
To calculate the total current through the device we only need to put in the wavefunction of the device
and the contacts (|
d
, |
1
, |
2
) from Eqs. 32, 34 and 9 and add all the contributions together. Thus the
current into the device from a incoming wave of one energy (E) in contact 1 (|
1,n
) through the coupling
dened by
2
is:
i
2 from 1
=
ie
h
_

2
|
2
|
d

d
|

2
|
2

_
(49)
=
ie
h
(
1,n
|
1
G

2
g

2
G
d

1
|
1,n

1,n
|
1
G

2
g
2

2
G
d

1
|
1,n
) (50)
=
ie
h

1,n
|
1
G

2
_
g

2
g
2
_

2
G
d

1
|
1,n
) (51)
=
e
h

1,n
|
1
G

2
G
d

1
|
1,n
(52)
7
Adding over the modes n and noting that the levels are lled from the reservoir connected to contact 1
gives (2 for spin):
I
2 from 1
= 2
e
h

_
E=
dE f(E,
1
)

n
(E E
n
)
1,n
|
1
G

2
G
d

1
|
1,n
(53)
=
2e
h

_
E=
dE f(E,
1
)

m,n
(E E
n
)
1,n
|
1
|mm|G

2
G
d

1
|
1,n
(54)
=
2e
h

_
E=
f(E,
1
)

m
m|G

2
G
d

1
_

n
(E E
n
)|
1,n

1,n
|
_

1
|m (55)
=
2e
h

_
E=
dE f(E,
1
)

m
m|G

2
G
d

1
a
1
2

1
|m (56)
=
e
h

_
E=
dE f(E,
1
)Tr
_
G

2
G
d

1
_
(57)
To get the total current through the device the current from contact two have to be subtracted away:
I =
e
h

_
E=
dE (f(E,
1
) f(E,
2
)) Tr
_
G

2
G
d

1
_
(58)
which is exactly the Landauer formula for the current.
Acknowledgments
The author is grateful for the inspiration from Supriyo Datta and Mads Brandbyge. The revised version
was initiated from discussions with Casper Krag and Carsten Rostgard.
References
[1] M. Paulsson, F. Zahid, and S. Datta. Nanoscience, Engineering and Technology Handbook, chapter
Resistance of a Molecule. CRC Press, 2002. Editors, D. Brenner, S. Lyshevski and G. Iafrate, cond-
mat/0208183.
[2] P. S. Damle, A. W. Ghosh, and S. Datta. Molecular nanoelectronics, chapter Theory of nanoscale
device modeling. 2002. Editor M. Reed.
[3] F. Zahid, M. Paulsson, and S. Datta. Advanced Semiconductors and Organic Nano-Techniques, chapter
Electrical Conduction through Molecules. Academic press, 2002. Editor H. Markoc.
[4] S. Datta. Electronic transport in mesoscopic systems. Cambridge University Press, Cambridge, UK,
1997.
8
[5] L. E. Ballentine. Quantum mechanics. World Scientic, 1998. see chapter 16, Scattering.
9