Fundamentals of Nanoelectronics

ECE495 - Session 7, Sept 9, 2009

Schrodinger Equation - Basic Concepts

Ref: Chapter 2.1

Professor Supriyo Datta

Class notes taken by: Mehdi Salmani

Review

During last classes we learned how

to calculate current for small device.
Given a set of energy levels or
Density of States (DOS), weve
talked about the I-V characteristics
of a device. What we want to
understand now is how to model
such energy levels / DOS and what is
the origination of these states.

VG

Gate
S
O
U
R
C
E

INSULATOR

D
R
A
I
N

UL= -qVG

VD

DOS

f (E)
1

2 qV

f (E)
2

Discrete Energy Spectrum

To achieve this, scientists started with the simplest material
possible, namely the Hydrogen atom.

q2
1
U=
+ mv 2 the first part is potential energy
4 0 r 2
and the second part is kinetic energy.

-q
+q

Fundamentals of Nanoelectronics, Session 7

Sept 9 2009
When they heated Hydrogen gas,
the results was discrete spectrum!

.
.
.
Original motivation for creating the
Schrdinger Equation comes from
discrete photon emission energy
bands that were observed when
Hydrogen-like atoms were heated.
These energy levels could not be
explained using classical physics.

hv

Schrdinger Equation

h2 2
v
E =
+ U (r )
2m

q2
and U =
4 0 r

There is only analytical solution for hydrogen atom and for all other atoms they solve it numerically.
And Schrodinger showed its analytical solution for Hydrogen.
How does the Schrdinger Equation lead to discrete energy levels?

Vibrating String

Vibrating String

The general one-dimensional wave equation is:

2
2u
2 u
=v
t 2
u 2

v is speed and u describes

displacement which is a function of time and position.

x =0
u~e

-it

=> u(x,t) = (x)e

2

-it

(-i) (x)e

-it

2
= v
((x)e-it)
2
u
2

x=L

Fundamentals of Nanoelectronics, Session 7

Sept 9 2009

-2 (x) = v 2

2
(x) This is time independent version
u 2

There are two types of solution for this equation:

1.
2.

Ae ikx or Ae ikx + Be ikx

A cos kx + B sin kx

If we use first type solution

-2 (x) = v 2

2
(x) =>
u 2

= v (ik ) = v k k =
2

2
v

k =

(I )

For boundary conditions (x = 0 and x = L) we will use the second type solution:
(x) = A cos kx +

B sin kx

and at x = 0 (x) = 0 then A = 0 => (x) = Bsin kx

at x = L (x) = 0 then B sin kx = 0 kL = n k =

( I ) & ( II ) = k =

=n
L
L

n
L

( II )

n is an integer value and the

is fixed then the

will get only discrete values.

f =

n
=
2 2 L

1-D Schrdinger Equation

h 2 2
v
E =
+ U (r )
2
2m x
h 2 2
v
ih
=
+ U (r )
2
t
2m x

Time independent and 1-D Schrodinger Equation.

Time dependent and 1-D Schrodinger Equation.

To achieve first from second equation we should use this equation: ( x, t ) = ( x)e iEt / h
The only difference between the 1-D Schrodinger eq. and guitar string eq. is the U(x) and this makes
the eq. solving hard because it is a function of x.
There are some assumptions to reduce the complexity and then solve the Schrodinger eq. like
assuming the particle in the box. While it is quantum analogue of a 1-dimensional vibrating string.
(the particle in a box (or infinite square well)).
Then the 1-dimensional Schrdinger Equation is: ih

h 2 2
=
t
2m x 2

Fundamentals of Nanoelectronics, Session 7

Sept 9 2009

h2k 2
= k
2m

Infinite
Square
Well

( x, t ) = e ikx e iEt h E =

Now Schrdinger equation is like guitar string

equation.

U=0

n
atx = 0 ( x) = B sin(kx) k =
L

x =0

h2 2 2

En =
n n = n
2
2m L
L

The last equations are called Dispersion relation. Because it makes relation between time
dependent part ( or E) and position dependent part (k).

x=L