Quantum Transport:

Atom to Transistor

Prof. Supriyo Datta

ECE 659
Purdue University

01.31.2003

Lecture 8: Schrödinger Equation: Examples

Ref. Chapter 2.3

Network for Computational Nanotechnology

 00:00
Retouch on Concepts

• Recall, the analytical form of the Summary:

Schrödinger Equation:
? - became a column vector
∂ r r
ih Ψ ( r , t ) = H op Ψ ( r , t ) U - became a diagonal NxN matrix
∂t
− h2 2 r − h2 2 - became a tri-diagonal NxN
where
H op = ∇ + U (r ) ∇ matrix
2m 2m

∂ • And the corresponding eigenvectors

• In numerical form: ih {? } = [Η ]{? } and eigenvalues form a set by which
∂t wave functions may be represented as a
With the eigenvalue equation [H]{a}= Ea {a} linear combination of eigen vectors.
{Ψ } = ∑ Cα {α }e − iE t hα

α
Last time we looked at the Schrödinger
Equation in one dimension. Along ‘x’ a ?n
discrete set of lattice points were placed.
With the discrete set of lattice points shown a
on the right, the Schr ödinger Equation was
converted into a difference equation. x
1 2 3 n-1 n n+1 N
 05:21
Retouch on Concepts

Finally, it is important to Computationally it is rather easy to find

emphasize that eigenvalues matrix eigenvectors and
correspond to energies and eigenvalues. For example in Matlab we
eigenvectors to wavefunctions . simply call “[V,D]=eig(H)”.

Wave Functions and Energies

Matrix D provides the eigenvalues in a diagonal
matrix and Matrix V provides the corresponding
eigenvectors as columns of a square matrix.

E3
D= V=
E2

E1
 08:39
Separating Variables

• The main advantage of the numerical • So, how do we go beyond the one
method is that once it is solved for a dimensional numerical solution seen thus far?
particular dimensional structure, that • In two dimensions the Hamiltonian operator
same method may then be applied with looks like:
little difficulty to other scenarios of the
h2  ∂ 2 ∂2 
same nature. Hop = −  2 + 2  + U ( x , y )
2m  ∂x ∂y 
Example: The slanted well has a The lattice for this operator is shown below .
complicated analytical solution
(involving Airy Functions) but
numerically the solution is the same as
that implemented for the infinite square It is straight forward
well. to extend the finite
difference method
to two dimensions,
Slanted Well the main problem is
size. i.e
100x100 pts=10,000 pts
U=0
 14:32
Reducing the Lattice

• One solution is to separate the 2 -D • Numerically this translates into two 1 -D

problem into two 1 -D. However, this is only lattices (one for the x-axis and one for the
possible if the potential is separable: y-axis).

U ( x , y ) = Ux ( x) + Uy ( y ) • For example, take a two dimensional

• Analytically this is done by separation of infinite square well:
variables.
Let, Ψ ( x, y) = X ( x ) × Y ( y ) 2-D Infinite Square Well
− h2 ∂2 
Thus, E xX ( x) =  + U x( x ) X ( x )
∂
2
 2 m x  y
 − h2 ∂2  Y(y)
E yY ( y ) =  + Uy ( y ) Y ( y )
 2 m ∂ x 2

Where E = Ex + Ey X(x)
From,
 − h2  ∂2 ∂2  
EΨ ( x, y ) =   2 + 2  + U ( x, y ) Ψ ( x, y )
 2m  ∂x ∂y  
time independent form 0 x
 22:00
2-D Infinite Square Well

• Numerically if we split the x and y lattices • Notice that two eigenvectors with the
by 100 points. How many eigenvectors will
same m+n value do not necessarily
result?
have the same energy:
• Take eigenvectors {? 22 } and {? 13 },
y Infinite Square Well which has more energy?
h 2π 2 n 2
Recall: En =
100 1 2mL2
2
1 ∴ (E13= E1+E3 ) > (E22= E2+E2)
2
1 2
• Now what about a 3-D infinite square
0 1 2 100 x
well or a 3-D box? The same thing,
separation of variables gives
•Ans: Total=100x100=10000 eigenvectors . eigenfunctions of
We combine X(x) and Y(y) functions such ? 111 , ? 112,….etc.
that
? nm (x,y)= X n (x) Y m (y)
 27:16
Hydrogen Atom

• Consider the Hydrogen Atom, can it be • Note: A similar approach can be

handled by separation of variables? taken with all atoms but not with
molecules because you don’t have
• The Hydrogen atom cannot be separated in this spherical symmetry.
Cartesian coordinates
• So separation of variables in
− h2  ∂2 ∂2 ∂2  spherical coordinates gives the radial
Hop =  2 + 2 + 2  Hamiltonian as:
2m  ∂x ∂y ∂z 
− h2   ∂ 2 2 ∂  L ( ?, φ ) 
− q2   2 + + 
2m   ∂r r ∂r  r2 
4pe0 x 2 + y 2 + z 2
−q2
But it can be separated in spherical 4 pe 0 r
coordinates such that U becomes
• In general ? becomes
− q2
4pe0r Ψ(r,?, f ) = R(r)Θ(?)Φ (φ )
 Time – 34:40
Hydrogen Atom

• Note: The radial equation • Hence, the entire radial equation

simplifies to:
− h2  d 2 2 d 
ER ( r ) =  2 +  R ( r )
2m  dr r dr  − h2 d 2 f (r) q2
Ef (r) = − f(r)
 − h 2 l (l + 1) q2  2m dr2 4πε0 r
+  − R(r)
 2m r
2
4 pe0r  h2 l(l +1)
+ f(r)
can be simplified by letting R(r) = f (r) /r 2m r 2
Thus, dR (r ) f ' ( r ) f (r ) • Furthermore, the radial equation gives
= − 2 the s,p,d …. levels of the hydrogen atom
dr r r for different values of “l”:
d 2 R (r ) f ′ (r ) f ′ (r ) f ′( r) 2 f (r ) s: l=0
= − − + p: l=1
dr 2 r r2 r2 r3
D: l=2,etc
 d2 2 d  f " (r )
 2 +  R(r ) =
∴  dr r dr  r
 40:54
Hydrogen Atom

• How do we get a numerical solution to the • Where,

hydrogen atom?
L(?,F )= 1 ∂  ∂Y 
• Ans: We set up a discrete axis along a radial  sin ? 
axis and solve the appropriate radial Hamiltonian
sin ? ∂?  ∂? 
matrix for l=0,1,2 …. 1 ∂ 2Y
+ ,
sin 2 ? ∂? 2

0 r
• Back to the hydrogen Hamiltonian, where does
l(l+1) come from?
•Ans: l(l+1) forms the eigenvalues of the
spherical harmonics or angular part of
Ψ (r, ?, f ) = R( r)Θ(?)Φ (φ )
and Yl m = Θ l( ?) Φ m(φ )
L(θ , φ )Yl m = −l (l + 1) Yl m
m
• Note: The angular term,Yl , is the
same for all atoms which results in
isotropic behavior. It is only the radial
term that changes.
 49:01
Summary

• Closing Notes:

• For higher dimensions use separation of variables whenever poss ible.

• Separable solutions are of the form Ψ (r, ?, f ) = R (r )Yl m (?, f )
•The angular term is the same for all separable spherical solutio ns.
• For hydrogen-like atoms the radial term can be treated as a 1-D numerical
problem with coulomb potential − Zq 2

4pe0r

h 2 l ( l + 1)
and angular “potential” +
2 mr 2