The Postulates of Quantum Mechanics

Some of the basic postulates of quantum mechanics will now be

introduced.

The postulates are the fundamental assumptions on which quantum

mechanics is based.

For example, Newton’s three laws are postulates of classical

mechanics.
 The Postulates of Quantum Mechanics
These postulates cannot be derived!
Newton’s Laws, are postulates and are believed to be correct
for macroscopic systems based on agreement with
experiment.
The postulates of Quantum Mechanics also can not be
derived. They too are believed to be correct for both
microscopic and macroscopic worlds based on agreement
with experiment.
The postulates of quantum mechanics can be expressed in many
ways.
People present them in different order and the number of
postulates is often different (some are combined)

The postulates are connected as a whole

They may be very strange at first. Applications will help us
understand.
 Classical State Function

The first postulate deals with a state function, so let’s examine the
classical state function.

The classical mechanical state of a one-particle system at a particular

time is given by 3 position coordinates and 3 velocities (or momenta)
The time evolution of the system is governed by Newton’s second law:
d 2x d 2y d 2z
Fx m 2 Fy m 2 Fz m 2
dt dt dt
Usually, given the initial positions and momenta, we can integrate
Newton’s equations of motion to obtain the whole trajectory of the
system and the state function. The the 3 coordinates and 3 velocities
completely define the system.

Heisenberg’s uncertainty principle hints that this is not valid because

we cannot simultaneously determine the position and momentum
exactly.
Postulate 1

The state of a quantum mechanical system is completely

specified by a function Y(x,y,z,t) that depends on the
coordinates and on time. This function is called the wave
function of the system. All possible information about the
system can be derived from Y.
Example of a Wave function
The wave function for a one-dimensional harmonic oscillator
is:
 x 2 i 2 Et
  
Y ( x, t )  Ce a
e h or

where a and C are constants and the variable x is the

displacement of the diatomic from equilibrium.

Notice the wave function can be complex (and complicated)!

More on the properties of the wave functions later.

For this idealized system, we can derive any property of the

system from the above function.
 Postulate 1 (extended)

The state of a quantum mechanical system is completely

specified by a function Y(x,y,z,t) that depends on the
coordinates and on the time. This function is called the wave
function of the system. All possible information about the
system can be derived from Y.
The wave function Y(x,y,z,t) has an important property that its
square is the probability distribution function for the system:

Y( x, y, z, t )  Y* ( x, y, z, t )Y( x, y, z, t )
2

Complex conjugate of the wave function is

Y ( x, y, z, t )
*
obtained by replacing any imaginary numbers
with their negative: i*  i
e.g.
 x 2  x 2 i 2 Et
i 2 Et   
Y * ( x, t )  C  e e
   a
Y ( x, t )  C  e e
a h h
The probability of finding the particle at a time t near the position
(x,y,z) is proportional to
P( x, y, z, t )  Y( x, y, z, t )  Y
2 2

More precisely, the probability of finding the particle at time t within a

given volume element dV = dxdydz is:

  Y( x, y, z, t ) dxdydz   Y*Ydxdydz
2

V V

The probabilistic interpretation of the wave function

was first developed by Max Born and it is sometimes
referred to as the Born interpretation of the wave
function.

Given this physical interpretation of the square of

the wave function, the wave function must satisfy
certain requirements. Max Born
Minimal Requirements for a Valid Wave Function
In general, Y(x,y,z,t) must satisfy the following conditions:

1. Y(x,y,z,t) must be a continuous function.

There cannot be points in space where the wave function is

“missing”. There must be a probability associated with every point in
space.

2. Y(x,y,z,t) must be single-valued and bounded.

• at a given x,y,z and t, Y can only have one value.
• the value of the wave function cannot be infinite anywhere.

There cannot be two or more probabilities of finding the particle at

a single point in space.
There cannot be an infinite probability of finding the particle at a
single point in space. The probability can at most be 1 which
corresponds to 100%.
Examples
Which of the following functions are acceptable wave functions
according to the first two criteria for the interval given?

sin x (, )
cos x
sin x  (, )
2

e  x cos x (0, )

ex (, )

(1  x 2 ) 1 (1, 1)
3. Y(x,y,z,t) should be normalizable such that the following integral
is equal to one:
  

   Y ( x, y, z, t )Y ( x, y, z, t )dxdydz  1
*



or

 Y Yd  1
*

all
space

Why?

The probability of finding the particle at an instant in time,

“somewhere” in the universe must be equal to 100% or unity.
Consider the following probability distribution function that describes
the probability of finding a single particle along the x coordinate:

P(x)

x1 x2
x

To find the probability that the particle will be between x1 and x2 we

would integrate P(x) between x1 and x2.
x2

 P( x)dx
x1

If we integrate over all space, then the probability must equal 1.

3. Y should be normalized such that the following integral is equal
to one:

 Y Yd  1
*

all
space

If the wave function is not normalized, it should at least be

normalizable such that the integral is equal to a real, finite value.

 Y Yd 
* finite and real
valued
all
space

If the above integral is finite and real valued, the wave function Y
can easily be normalized by multiplying by a normalization constant.
  Y Yd 
* finite and real-
valued
all
space

If the above is true then the normalization constant can be

determined from the following:
1
  2

N  
1
 Y *Yd 
 

And our normalized wave function is then:

Ynormalized  N  Y
Examples
Normalize the following functions, if possible, for the intervals given.

sin x (, )

sin x (0,2 )

xe  x (0, )

e x (, )
2
 4. Y(x,y,z,t) should have partial derivatives

Y Y Y
x y z
that are continuous functions of x, y, z.

This is a looser requirement for a ‘good’ wave function. There are

sometimes cases when this is relaxed. For example, when potential
is ill-behaved. (e.g. Coulomb potential at the nucleus is infinite)

Conditions 1 through 4 are the requirements for a ‘well-

behaved’ function.
 Minimal Requirements for a Valid Wave Function

1. Y(x,y,z,t) must be a continuous function.

2. Y(x,y,z,t) must be bounded and single-valued.

3. Y(x,y,z,t) must be normalizable.

4. Y(x,y,z,t) should be smooth. (The derivatives should be

continuous).

These requirements arise from the Born-interpretation of the wave

function. It is important to understand why they arise from the Born
interpretation of the wave function.
Postulate 2

To every “observable” in classical mechanics there

corresponds an operator in quantum mechanics.
More to come

What is an operator?
An operator is a mathematical entity or symbol that tells you
to do something to whatever follows the symbol.

ˆ d
let O and f ( x)  x 2
dx
ˆ d 2
Then Of ( x)  ( x )  2 x
dx
Operators usually denoted with a carat ^ over it, e.g. Ô
In these lecture notes will sometimes use (in the text) a bold
underscore, e.g. O
Postulate 2 (Complete)

To every observable in classical mechanics there

corresponds an operator in quantum mechanics. To find the
operator, write down the classical-mechanical expression for
the observable in terms of Cartesian coordinates and linear
momentum, and make the following replacements:

x  xˆ  x px  pˆ x  i
x
y  yˆ  y 
p y  pˆ y  i
z  zˆ  z y

t  tˆ  t pz  pˆ z  i
z

h is the reduced Planck’s constant


2 and is pronounced ‘h-bar’
Define the kinetic energy operator in one-dimension.

Define the kinetic energy operator in 3-dimensions.

The first postulate states that the wave function of a quantum
mechanical system contains all information concerning the system.

The second postulate states that for every physical observable, there
corresponds a quantum mechanical operator!

The 3rd postulate deals with measurements in quantum mechanics.

Postulate 3

Any measurement of the observable associated with the

operator A, the only values that will ever be observed are the
eigenvalues ‘a’, which satisfy the following:
Âf af

f is an eigenfunction of the operator A

(It is not necessarily the wave function)
 What are Eigenfunctions and Eigenvalues?
When a mathematical operation (such as multiplication, differentiation)
is performed on a function, the result is generally some different
function.
For example: differentiation of x2 yields a different function, 2x.

For some combinations of operations and functions, the same function is

regenerated, multiplied by a constant.
For example: differentiation of e2x yields a different function, 2e2x.
The original function is simply multiplied by 2.

Aˆ f ( x)  a  f ( x) It gives the exact

function back
Operator acts on multiplied by some
function Y constant

The function is called an eigenfunction and the numeric constant is

called the eigenvalue. The above is called an eigenvalue equation.
 Determine if the function f is an eigenfunction of the operator A
(if so, what is the eigenvalue?)

ˆ d
A f (x ) cos( x )
dx

ˆ d2 f (x ) cos( x )
A
dx 2
d2
ˆ
A f (x ) ei x
dx 2

ˆ d Ax 2
A f (x ) e
dx
ˆ 
A f (x, y) y 2e 6x
y
 Postulate 3
For any measurement of the observable associated with the
operator A, the only values that will ever be observed are the
eigenvalues an which satisfy:
ˆ
Af an fn
n

An operator generally has more than one valid eigenfunction so we index

the eigenfunctions with a subscript n.
The eigenvalue ‘a’ can be discrete or continuous, depending on the
operator.
If the eigenvalues are discrete then we usually index them from the smallest
value to the largest value:
ˆ
Af an fn
n n = (sometimes) 0,1, 2, 3….
If the eigenvalues are continuous, we usually drop the subscript.

Âf af
In any experiment measuring the observable corresponding to the
operator A, only the values a1, a2, a3,… will be observed.

Observables of the Linear Momentum, px

Consider the linear momentum, px, which is an observable we can often
easily measure for a given single-particle system.
By the third postulate, we will only be able to measure values that are
eigenvalues of the following equation:

pˆx f af

 i f  af
x
It turns out that eikx is an eigenfunction of the operator px. Check that
this is true.
 a  k
since k can be any constant, it turns out that linear momentum is an
observable that is continuous and can take on any value.

Observables of the Angular Momentum, Lz

The angular momentum operator (in the z-direction) is represented by
the following operator in quantum mechanics:

ˆ    
Lz  i x  y 
 y x 

The eigenfunctions the angular momentum is given the symbol Y.

Lˆ zY  mY m = 0, ±1, ±2, ±3 …

Don’t worry about the explicit form of Y at this moment.

 a  m m = 0, ±1, ±2, ±3 …

Notice, however, that the constant ‘m’ can only take on discrete integer
values. Thus, the eigenvalues of the angular momentum operator can
only assume discrete values.
When we go to measure the angular momentum of any system, we will
only be able to measure these discrete values.

It turns out that angular momentum is quantized. (More on the

quantization of angular momentum much later in the course)
 Postulate 3
For any measurement of the observable associated with the
operator A, the only values that will ever be observed are the
eigenvalues an, which satisfy:
ˆ
Af an fn
n

Postulate 3 tells us what values of a given observable are allowable.

But what value will we actually measure?

If we have a system described by the wave function Y, the first postulate

tells us that any information we can derive about the system is contained
in the wave function.

Unfortunately, due to the probabilistic nature of quantum mechanics, we

can’t be sure which eigenvalue we will actually measure (a1, a5, a192?)

This is quite different from classical mechanics where if we knew the

state function x(t), we could tell exactly what one would measure at any
given time.
 Young’s Double-Slit Experiments with
Single Photons

Interference Pattern Made Up of Discrete Particle Detections!

 Postulate 4
If a system is in a state described by a normalized wave
function Y, then the average or mean value of the
observable that will be measured corresponding to the
operator A is given by:

 a   Y Aˆ Yd *

all
space

The above is called the expectation value in quantum

mechanics.

If we have a wave function and an operator corresponding to an

observable we want to measure, the above tells us what the average
value we will measure at a given time.
 Let’s write this out more explicitly:

 
  
ˆ Y ( x, y, z, t ) dxdydz
 a     Y * ( x, y, z, t ) A


Brackets are used to emphasize that we apply the operator to Y

before we integrate.

If the wave function Y is not normalized, then the expectation value

is given by:

 
  

 Y ( x , y*
, z , t ) ˆ
A Y ( x, y, z, t ) dxdydz

  

   Y ( x, y, z, t )Y ( x, y, z, t )dxdydz
*


UNTIL NOW….
We have discussed some properties of the wave function Y. However,
we have not discussed how we determine the wave function. The next
postulate, which involves the famous Schrödinger equation, gives us
this.

Erwin Schrödinger (1887-1961)

Postulate 5

The wave function of a system evolves in time according to

the time-dependent Schrödinger equation:
 
HY( x, y, z, t )  i Y( x, y, z, t )
t
Where H is the Hamiltonian operator given by:
  2
H  Tˆ  Vˆ    2  V ( x, y, z, t )
2m

We can use the Schrödinger equation to solve for the wave

function for any system. (in principle).
Postulate 5 (alternative expression)

The wave function Y(x,y,z;t) of a system can be obtained by

solving the time-dependent Schrödinger equation for that
system:
 
HY( x, y, z, t )  i Y( x, y, z, t )
t
where
  2
H  Tˆ  Vˆ    2  V ( x, y, z, t )
2m

The potential energy function V cannot be given in general

because it depends on the system.
In this course, the systems we are interested in usually have a
potential which does not depend on time.
V ( x, y, z, t )  V ( x, y, z)
In this case, we can simplify things.

It turns out that if the Hamiltonian (or the potential within it)
does not change in time we can write the wave function as:

Y( x, y, z, t )   ( x, y, z)  f (t ) Separation of variables
 Addition to Postulate 5

If the potential does not change in time, the wave function

can be obtained by solving for Schrödinger’s time-
independent equation.

Hˆ  ( x, y, z )  E ( x, y, z )
With the total wave function given by:
iEt
Y( x, y, z, t )   ( x, y, z )  e 

Thus, we don’t have to solve the time-dependent wave equation, only

the time-independent equation, because the time-dependent component
of the total wave function is always the same.

We will use Schrödinger’s time independent equation extensively in this

course.
 Stationary State Wave Function
When the Hamiltonian does not change in time, the solutions to the time-
independent wave equation are called stationary states.
iEt
Y( x, y, z, t )   ( x, y, z )  e 

The reason for this is that the probability density does not change in
time, i.e. it is time independent.
This is easily shown (in 1-D):
P( x, t )  Y( x, t )  Y( x, t )* Y( x, t )
2

*
  ( x)  e   ( x)  e iEt 
iEt

 
iEt iEt
  ( x)  ( x)  e
*
e  

  ( x)  ( x)  e
* 0

  ( x)  ( x)
* independent of time
Notice that the time component of the total wave function of a stationary
state simply oscillates:
iEt
Y( x, y, z, t )   ( x, y, z )  e 

This is easier to see if we let: w  E /   E /( 2h)  / 2

wt
Y( x, y, z, t )   ( x, y, z)  e
  ( x, y, z)  coswt   i sinwt 
w is then the angular frequency of this oscillation.

Because Schrödinger’s equation has wave-like solutions, this is the

reason Schrödinger’s formulation of quantum mechanics is often called
wave mechanics.
 Summary of the Postulates of Quantum Mechanics

Postulate 1
The state of a quantum mechanical system is completely
specified by a function Y(x,y,z,t) that depends on the
coordinates of the particle and on time. This function is
called the wave function of the system. All possible
information about the system can be derived from Y.

The wave function Y(x,y,z,t) has an important property that its

square is the probability distribution function for the system:
Y( x, yz , t )  Y* ( x, y, z, t )Y( x, y, z, t )
2

Due to the Born interpretation of the wave function Y(x,y,z,t) must be

‘well behaved’. It must be a continuous function that is bound and single
valued everywhere. Y(x,y,z,t) must be normalizable or quadratically
integrable. Finally, the partial derivatives of Y w.r.t. to x, y and z should
be continuous.
Postulate 2

To every observable in classical mechanics there

corresponds a linear Hermitian operator in quantum
mechanics. To find the operator, write down the classical-
mechanical expression for the observable in terms of
Cartesian coordinates and linear momentum, and make the
following replacements:

x  xˆ  x px  pˆ x  i
x
y  yˆ  y 
p y  p y  i
ˆ
z  zˆ  z y

t  tˆ  t pz  pˆ z  i
z

This postulate gives us a simple recipe for obtaining quantum

mechanical observables.
Postulate 3
Any measurement of the observable associated with the
operator A, the only values that will ever be observed are the
eigenvalues a, which satisfy:
Aˆ n  ann

Postulate 4

If a system is in a state described by a normalized

wavefunction Y, then the average or mean value of the
observable corresponding to operator A is given by:
* ˆ
a   AYd
Y
all
space
The above also called the expectation value in QM.
 Postulate 5

The wave function of a system evolves in time according to the time-

dependent Schrödinger equation:
ˆ 
HY( x, y, z; t )  i Y( x, y, z; t )
t
Where H is the Hamiltonian operator given by:

 2
Hˆ  Tˆ  Vˆ    2  V ( x, y, z; t )
2m
If the potential does not change in time, the wave function can be
obtained by solving for Schrodinger’s time independent equation.

Hˆ  ( x, y, z )  E ( x, y, z )
Here, the the total wave function given by:
iEt
Y( x, y, z, t )   ( x, y, z )  e 
• The postulates of quantum mechanics replace Newton’s and
Hamilton’s equations of motion for describing the behaviour of a
given system.

• Postulate 5 furnishes the “equation of motion” that yields the wave

function describing the possible states of the system.

• Once Y(x,y,z;t) are obtained - ALL of the properties (physical

quantities) of the system can be calculated and interpreted by
applying the other postulates.

• We will obtain an understanding of quantum mechanics by

applying the postulates to a variety of chemical systems.