Expectation (or mean) value for position

ò
-¥
dx |y ( x, t ) |2 = 1

Using the probability interpretation,

¥
< xˆ > = ò
-¥
dx x | y |2

¥
= ò dx y *
-¥
x y

1
Probability

= 14

2
Probability – Most probable value

= 14

3
Probability – Median

= 14

4
Probability – Average or Expectation Value

= 14

This is called the

Expectation Value in QM
5
Probability – Average of squares

= 14

6
Probability – Variance or Standard Deviation

7
Continuous Variable…

8
Expectation value: Statistical interpretation

Keep measuring position of the particle many times

and average the result.

Statistical Ensemble
-Consider large number of similarly prepared systems
-Make measurement on each of them
-Average all the measurements

9
Expectation value for position spread
¥
< xˆ > = ò dx y * x y
-¥
¥
< xˆ 2 > = ò
-¥
dx y * x 2 y

Dx = < xˆ 2 > - < xˆ > 2

10
Illustration:
y = A exp[ - a x2 ]
¥
< xˆ > = ò dx y * x y
-¥
= 0

¥
< xˆ 2 > = ò
-¥
dx y * x 2 y > 0

11
 Time derivative of < x̂ >
¥
d < xˆ > é ¶y * ¶y ù
dt
= ò
-¥
dx ê
ë ¶ t
x y + y*x
¶ t
ú
û

Use Schrödinger equation and its complex conjugate

¶y i ! ¶2 y i
= - Vy
¶t 2m ¶ x 2
!

¶y * i ! ¶2 y * i
= - + Vy*
¶t 2m ¶ x 2
!
12
Time derivative of < x̂ >continued …
¥
d < xˆ > é ¶y * ¶y ù
dt
= ò
-¥
dxê
ë ¶ t
x y + y*x
¶ t
ú
û
¥
æ -i ! ö é ¶2 y * ¶ 2y ù
= ç
è 2 m ò
÷ dx x ê
ø -¥ ê
ë ¶ x 2
y -y *
¶ x 2 ú
úû
¥
æ -i ! ö ¶ é ¶y * ¶y ù
= ç
è 2 m ò
÷ dxx
ø -¥ ¶ x
ê
ë ¶ x
y -y *
¶ x
ú
û

Integrate by parts …
13
Time derivative of < x̂ >continued …
ì é ¶y *
¥ ü
ïx ¶y ù ï
ê y -y * ú
d < xˆ > æ -i ! ö ï
ï ë ¶ x ¶ x û-¥
ï
ï
= ç ÷í ý
dt 2 m ¥
è øï é ¶y * ¶y ù ï
ï
ï
î
- ò
-¥
dx ê
ë ¶x
y -y * úï
¶ x ûï
þ
¥

ò
-¥
dx |y ( x, t ) |2 = 1

y ( x, t ) ® 0 faster than 1/ x
14
Time derivative of < x̂ >continued …
¥
é ¶y * ¶y ù
x ê y -y * ú ® 0
ë ¶ x ¶ x û -¥

¥
d < xˆ > æ -i ! ö é ¶y ¶y * ù
dt
= ç
è 2 m ò
÷ dxê
ø -¥ ë ¶ x
y * -y
¶ x
ú
û

Integrate second term by parts …

15
Time derivative of < x̂ >continued …
¥ ¥
¶y * ¶y
ò ò
¥
dx y = y y * -¥ - dxy*
¶x ¶x
-¥ -¥

¥
d < xˆ > æ ¶ ö
< pˆ > = m
dt
= ò
-¥
d x y * ç -i !
è
÷y
¶xø
¥
OR < pˆ > = ò
-¥
d x y * pˆ y

16
Expectation value of an operator
¥ ¥
< xˆ > = ò d x y * xˆ y
-¥
< pˆ > = ò d x y * pˆ y
-¥

General operator as a function of x and p operators

ˆ ˆ ¶
Q = Q ( x , - i! )
¶x
¥
< Qˆ > = ò
-¥
d x y * Qˆ y

17
Operator algebra
(sum and product of operators)

Action of an operator on y ( x, t ) produces another

wavefunction in general

Qˆ y ( x, t ) = f ( x, t )
Addition of two operators
(Qˆ1 + Qˆ 2 ) y ( x, t ) = Qˆ1y ( x, t ) + Qˆ 2 y ( x, t )

Qˆ1 + Qˆ 2 = Qˆ 2 + Qˆ1
18
Operator multiplication is not always commutative

Qˆ1 Qˆ 2 y ¹ Qˆ 2 Qˆ1 y

Illustration: ¶y
xˆ pˆ y = - i ! x
¶x
¶ (x y ) ¶y
pˆ xˆy = - i ! = - i! ( y + x )
¶x ¶x
( xˆ pˆ - pˆ xˆ ) y = i ! y
Fundamental
[ xˆ , pˆ ] = i ! commutation
relation
19
Scalar product of wavefunctions
¥
Define: (f , y ) = ò dx f * ( x) y ( x)
-¥

( f , y ) = ( y , f )*

(y , y ) ³ 0
If ( f , y ) = 0 , the two wavefunctions
are orthogonal.

Expectation value: < Qˆ > = ( y , Qˆ y )

20
Adjoint of an operator: Q̂ †

Definition: ( Qˆ † f , y ) = ( f , Qˆ y )

Illustration: Consider operator ¶

Ô =
¶x
¥
¶ ¶
(f,
¶x
y ) = ò
-¥
dx f*
¶x
y

¥
¶
= f *y |¥
-¥ -
-¥
ò dx
¶x
f *y
21
Adjoint of an operator continued …
¥ *
¶ æ ¶ ö
(f,
¶x
y ) =
-¥
ò
d x ç-
è
f÷ y
¶x ø
¶
= (- f, y )
¶x
¶ † ¶
( ) = -
¶x ¶x
¶ † ¶
Also, (-i ! ) = -i!
¶x ¶x
22
Self-adjoint or Hermitian operator (definition):

Qˆ † = Qˆ

Expectation value of self-adjoint operator

< Qˆ > = ( y , Qˆ y ) = ( Qˆ † y , y ) = (Qˆ y , y )

< Qˆ > * = ( y , Qˆ y ) * = (Qˆ y , y )

Observable quantities correspond to

Hermitian operators
23
Eigenvalues and eigenvectors of matrix A

Au = a u

(A- a I ) u = 0

For this system to have a non-zero solution,

det ( A - a I ) = 0

Polynomial equation in the eigenvalue “a”

24
Illustration: æ1 2 ö
A = ç ÷
è 2 4ø
Eigenvalues: (1 - a)(4 - a) - 4 = 0
a = 0, 5

æ 2ö æ1 ö
Eigenvectors: u1 = ç ÷ , u2 = ç ÷
è -1 ø è 2ø

25
Eigenvalues of Hermitian operator Q = Qˆ
ˆ †

Suppose that a set of wavefunctions {y n }

is such that:
Qˆ y n = qn y n

Eigenfunction Eigenvalue (number)

We have:
( y m , Qˆ y n ) = qn ( y m , y n )
( Qˆ y m , y n ) = qm * ( y m , y n )
The L.H.S.’s are equal for Hermitian operator
26
Eigenvalues of Hermitian operator … continued

(qn - qm *) ( y m , y n ) = 0

m = n Eigenvalues are real

m ¹ n Eigenfunctions belonging to distinct

eigenvalues are orthogonal

Why is a set of orthogonal vectors important ?

27
 Expansion of a general state in terms of
eigenfunctions of Q

y = åc
n
n yn

Normalization of wavefunction
(y , y ) = 1
= åå c
n m
n * cm (y n , y m )

= å
m
| cm |2

28
Expectation value of Q for the state y
< Qˆ > = ( y , Qˆ y )

å
= (
n
cn y n , Qˆ åc
m
m ym )

= åå c n * cm qm ( y n , y m )
n m

= å m
| cm |2 qm

29
Wavefunction collapse to eigenstate
(Bohr, Heisenberg)
On one measurement of Q, the system wavefunction
collapses to one of the eigenfunctions of Q. The measured
value of Q is the corresponding eigenvalue.

The probability of collapsing to state n is |cn|2

The expectation value of Q is weighted average of the
eigenvalues

30
Interaction free bomb detection J
( Elitzur and Vaidman 1993)
Mach-Zehnder interferometer Two paths
out of phase
by p
Mirror

Two paths
Beamsplitter in phase

Laser

Mirror
Beamsplitter

31
Bomb in one path: Like closing one slit

Mirror

Beamsplitter

Mirror
Beamsplitter

32
Interaction free bomb detection

50 % bombs explode

25 % bombs are detected without any interaction

25 % remain unknown

If we keep repeating the process with unknowns,

the total number detected without interaction ~ 33%

33
Problem 1: Uncertainty and diffraction spread

2
æ sin a ö
I = I0 ç ÷
Show that the diffraction è a ø
spread can be understood
æ p a sin q ö
in terms of uncertainty relation.
a = ç ÷
è l ø
34
Position uncertainty ~ slit size

Dx ~ a
Momentum uncertainty h
Dp ~
a
Position uncertainty in detecting photon on screen
Dp h l lD
D xscreen ~ D ~ D =
p a h a

Central lobe of diffraction pattern l

sin q1 =
(first zero of pattern) a
35
Problem 2: Eisberg-Resnick problem 34 (Chapter 3)

Drop a marble from height H

What is the average distance

by which you will miss the line
directly below?
H

36
36
Problem 2: Eisberg-Resnick problem 34 (Chapter 3)

Initial uncertainty in position of marble

Dx ~ a
Corresponding uncertainty in momentum of marble

!
Dp ~
a
As the marble falls from height, its position uncertainty is:

Dp !
Dx ~ a + t = a + t
m ma
37
 Time required to fall through height H :

2H
t =
g

Position uncertainty at the bottom of the ladder:

! 2H
Dx ~ a +
ma g

Depending on “a” there is a minimum for position uncertainty

38
Take derivative with respect to “a”

d ! 2H
( D x) = 1 - = 0
da m a2 g

! 2H
a =
m g

This is the optimal initial uncertainty in position

39
Final position uncertainty for the optimal “a”

! 2H ! 2H
Dx ~ a + ~ 2
ma g m g

Factors of 2 are not important.

!
You will lose them if you use D xD p ~
2
40
Problem 3: Uncertainty and size of hydrogen atom

Use uncertainty principle to find the radius of

Hydrogen atom in ground state.

Hint:
Assume some average radius “a”.

Total energy of electron = Kinetic energy +

potential energy (Coloum potential)

Minimize the total energy w.r.t. “a”.

41
3. Uncertainty and size of atom

Uncertainty of position for electron ~ atomic size

Dx ~ a
!
Uncertainty of momentum for electron Dp ~
a
Dp ~ < p 2 > - < p >2

Assume symmetric electron cloud

< p> = 0
42
Kinetic energy of electron

< p2 > < (Dp)2 > !2

< EK > = = =
2m 2m 2m a 2

Potential energy of electron

e2
< EP > = -
4p e0 a

43
Total energy

!2 e2
<E> = -
2m a 2
4p e0 a

Find value of radius a that minimizes total energy

d !2 e2
<E> = - + = 0
da ma 3
4p e0 a 2

4 p e 0 !2
a = 2
Bohr radius
me
44
Problem 4: s – orbital wavefunction for Hydrogen
is given by:

y = A exp( - r / a )

Calculate the constant A

Calculate the expected value of radius “r”.

Hint: Note that all integrals are now on 3D space

45
y = A exp( - r / a )
ò dV |y |2 = 1

¥
1 = A2 4p ò
0
r 2 dr exp( - 2r / a )

1
A =
p a3
¥
3
ò
2 3
<r> = A 4p r dr exp( - 2r / a ) = a
2
0

46
Problem 5: A particle in a deep potential well
has wavefunction at some instance given by

V ®¥
V ®¥

3p x
y = A sin( ) Energy
L

x
L
Find the value of A
Find the probability of finding the particle in
the interval between x = L/3 and x = 2L/3.
47
 L L 2
é 3p x ù
1 = ò0
d x | y |2 = A2 ò
0
d x êsin(
ë L û
)ú

2
A =
L

2 L /3 2
é 3p x ù
Probability = A2 ò
L /3
d x êsin(
ë L û
)ú

48
Problem 6: Show that

( xˆ 2 pˆ - pˆ xˆ 2 ) y = 2i ! xy

¶ ¶
( xˆ 2 pˆ - pˆ xˆ 2 ) y = - i! ( x 2 y - ( x 2y ) )
¶x ¶x

49
 Problem 7: Prove Ehrenfest theorem:

d ¶V
< pˆ > = < - >
dt ¶x

Hint:
Use calculations similar to the ones done in class for

d
< xˆ >
dt

50
d ¶y * ¶ ¶ ¶
dt
ˆ
ò
< p > = - i! dx [
¶t ¶ x
y + y*
¶ x ¶t
y]

¶y i ! ¶2 y i
= - Vy
¶t 2m ¶ x 2
!

¶y * i ! ¶2 y * i
= - + Vy*
¶t 2m ¶ x 2
!

51
Problem 8: The wavefunction of a particle is given
at time t = 0 by

y = A ( x 2 - a 2 ), | x | £ a
= 0 , | x| > a

- Find the constant A

- Find < xˆ > , < pˆ >, < xˆ 2 >, < pˆ 2 >, D x , D p

52
Problem 9: Angular momentum operator is defined as

Lˆ x = yˆ pˆ z - zˆ pˆ y

similar definitions for Lˆ y , Lˆ z

Show that Lˆ x Lˆ y - Lˆ y Lˆ x = i ! Lˆ z

53
Problem 10: Find the adjoint operators corresponding to

¶ ¶2
,
¶ x ¶ x2

54