UNIT 7 OBSERVABLES AND OPE ORS

Structure
7.I Introduction
Objectives
7.2 Quantum Mechanical Operators
Properties of Operators
Expectation Values
7.3 Eigenfunctions and Eigenvalues
7.4 Ehrenfest theorem
7.5 Summary
7.6 Terminal Questions
7.7 Solutions and Answers

7.1 INTRODUCTION
In thc block introduction we had said that quantum mechanics developed along two tracks.
In Units 4 to 6 we have presented one of these. You have learnt about the wave-particle
duality in Unit 4, In the previous unit, you have studied how Erwin Schrodinger discovered
the wave equation for matter waves, which is now called the SchrCidinger equation. You
have also learnt that in quantum mechanics any state of a systenl is represented by a wave
function y~ which can be obtained by solving the Schrtidinger equation. You have
studied the probabilistic interpretation of given by Max Born, that the de-Broglie
Schrijdinger waves are waves of probability which also satisfy the uncertainty relation.
Thus Units 4 to 6 present the wave nteclzanics version of quantum mechanics. You should,,
of course, realise that the existence of yf and the ScllriSdinger equation which form the
basis of wave mechanics are postulates. Thus quantum mechanics is based on certain
postulates which are not proved' but are like axioms of geometry.

In this unit, we will i~ltroducethe other track of the development of quantum

mechanics which is known as matrix mechanics and was developed by Werner
Heisenberg, Max Born and P. Jordan in the years 1925 and 1926. In this approach, only
physically observable quantities appear. Each physical quantity is associated with an
'operator' which can be represented by a matrix. What is an operator? You will find'
an answer to this question in the next section. You will learn the method of converting
a classical function into a quantum mechanical operator. This method is also one of the
postulates of quantum mechanics. The essential difference between classical mechanics
and matrix mechanics version of quantum mechanics is this: quantum mechanical
operators obey non-commutative algebra. We will explain what this means in
Sec, 7.2.2 and briefly discuss commutator algebra along with some of its applications.

The two tracks (Schrodinger's wave mechanics and Heisenberg's matrix mechanics)
were integrated by Paul A.M. Dirac who invented an abstract formalism for quantum
mechanics in 1930. In the remaining unit we shall present some basic concepts of this
unified formuiation of quantum mechanics given by Dirac (see Fig. 7.1). One of the
basic postulates in this formalism connects the measured value of a dynamical variable
with its theoretical value obtained with the help of the wave function yr. We introduce
it in Sec. 7.2.2. Thus we shall be able to relate quantum mechanical operators to
physically observable quantities.

In certain circumstances, it is possible that when an operator operates on a wave

function, the result may be a multiple of the same wave function. This gives rise
to what we call the eigenvalue-eigenfunction equation which you will study in
Sec. 7.3. Finally, you will learn about the Ehrenfest theorem which shows the
similarity as well as one of the basic differences between classical and quantum
mechanics.

The concepts presented in this unit may appear too mathematical and abstract to you in
the first reading. However, the formalism presented here is a very powerful and elegant
52 way of working with cluanturn rneclianical systems.
 Objectives '. Observables and Operaton

After studying this unit you should be able to

express a classical dynainical variable as a quantum mechanical operator,

o define the herhitian operator and the parity operator and apply their properties to
quantum mechanical systems,

0 compute h e expectation value of an operator,

derive elementary results of commutator algebra,

0 calculate the eigenvalues and eigenfunctions of a given operator,

o derive and interpret Ehrenfest theorem.

7.2 QUANTUM ,MECHANICAL OPERATORS

What is a quantum mechanical 'operator'? Let us begin with an analogy to explain this
idea. When you exercise, your muscles build up by the action of the exercise: Exercise Fig. 7.1 : Paul A.M. Dirac,
changes the muscles. Thc action of quantum mechanical operators on functions is a bit English physicist.
He was one of the
like that of exercise on muscles: they change the functions. You know from classical
pioneers OF
mechanics that the dynamical state of a system is determined at each instant of time by qllantum mechanics.
the knowledge of certain physical quantities, such as the position, velocity, linear Re also farmuiatcd
momentum, angular momentum, energy etc. of the particles constituting the system. thc relativistic wave
equatlon of
These physical quantities are also called dynamical variables. quantilm mechnnlw
which predicted the
The dynamical variables associated with a system can be measiired and provide exlstence of
information about the system at a particular point in space-time. In quantum mechanics, pr#ritron. Hc wa8
' all dyrzatnical variables are represented by operators because they bring about changes awarded thc Nobel
in the wave functions upon which they act. Prize in 1933.

All measurable attributes of a quan~ummechanical system are called observables, and

yet another postulate of quantum mechanics states that

Every physical observable is associated with a n operator Pustulate 4 :

which acts o n the wave function, Description of physical
quantities
Most of these dynamical variables or observables are functions of position (x), linear
momentum (p) and time (r) variables. Thus, if we can represent x and p by operators
we shall be able to express most of the remaining dynamical variables as operators. A
method to convert these variables into quantum mechanical operators is postulated in
quantum mechanics as follows:

i.e., when operator x,,, operates upon ly, the nsult is simply the multiplication of yr
by the variable x. In other words, the operator corresponding to the dynamical variable
x is x itself.

Thus, the momentum operator acting on the wave function results in its differentiation
with respect to the conjugate position coordinate x and the result is multiplied by -iR
Thus, the operator sf p, is -iha/a~.

Remember that Eqs. (7.1) and (7.2b) are postulated, i,e,, they can't be proved. The
An Intr@duc(lonto Quantum position and momentum operators are used in the construction of operators of other
Mccbenlcs dynamical variables such as angular momentum, energy etc. How do we do this? For
this, we take the classical expression for any operator D in terms of x and p, and use
Eqs. (7.1 and 7.2b) to obtain its operator form DO,, (x, -iha/ax, t). Notice that the time
variable has been retained as itself in the operator formalism. In quantum mechanics,
time is not treated as an operator. It is a dynamical variable.
To understand this method further, consider the example of the kinetic energy of a free
particle of mass rn executing one-dimensional motion given by p:/2rn. The quantum
mechanical operator of the kinetic energy is obtained by replacing p, by (p,),,,. Then
using Eq. (7.2b) we get

Furthermore, if its potential energy is given by the function V(x), then its potential
energy operator will also be V(x) since xop is x itself:

Now do you notice that the sum (-- 'L+

2m 2 2
V (XI)",,is nothing but the quantum
mechanical operator of the Hamiltonian which appears in the Schrodinger equation?

Thus, we obtain the Hamiltonian operator:

In this manner, you can convert most of the dynamical variables into quantum
mechanical operators. An important quantum mechanical operator that you will
encounter in our subsequent discussions is the parity operator. We will introduce it
here.
The Parity Operator
Parity is a simple but very useful concept in quantum mechanics. Consider a wave
function y(x). If on changing x to -x, the following relationship is obtained

then we say that the function y(x) has a definite parity. If y(-x) = t y(x), then y(x) is
of even parity. On the other hand, for y(-x) = - y(x), the parity of ~ ( x is) said to be
odd. All functions which do not obey (7.5) are said to have mixed parity. The parity
operation is equivalent to transforming a right-handed system of coordinates into a left-
handed one. Do you recall where you have first encountered this operation? It was
introduced in Unit 1 of the elective PHE-04 (Mathematical Methods in Physics-I) where
t operation in relation to vectors. The parity operator is
you have studied a b ~ u this
defined by

You can readily see that the parity operator is a space inversion operator, i.e., under its
operation x + -x, Thus if y(x) describes the state of a system, P y(x) describes its
mirror image.
We have discussed a method to obtain quantum mechanical operators from the
corresponding classical expression by changing x by x, t by t and p, by - --
iA
a
ax
However, there is no classical expression, in terms of x and p, which changes the sign
of the argument of a function by its operation. Hence, we say that the parity operator
, has no classical analog.
You should now carry out a couple of quick exercises to fix all the ideas presented so Observablcs and Operators
far in your mind.

SAQ 1 Spend
10 min
(a) Express the variables py and p, in operator form.

(b) Write the three components of the angular momentum L in terms of x , y, z and p,,
py and p, and thus obtain quantum mechanical operators for L,, L,, and L,.

You have just studied that in quantum mechanics, the measurable classical dynamical
variables like position, momentum etc. are represented by operators. These operators act
on a wave function and change it in some way. We have surnmarised the results
obtained so far for ready reference.

Dynamical Variables and Corresponding Operators

Dynamical Variable Operator

Position coordinate x x

x component of momentum px a
-iA -
ax
p,'
Kinetic energy T = - _A2 2
2n1 2m a2
Potential energy V (x, t) V ( x , t)
2
px --
Total energy -2m
+ V (x, t) Hamiltonian H = " + V(x)
2m 32-

Angular Momentum

Lx YPz- ZP,

$1 ZPx - XPz
4 XP,, - YP,
-
Let us now discuss some iinportant properties of these operators.

7.2.1 . Properties of Operators

Firstly, operators in quantum mechanics are generally linear operators. What is a linear
operator? By definition,' a linear operator satisfies the following properties:

where c is an arbitrary complex number. In general, we can combine Eqs. (7.7a and b)
and write for a linear operator:

where h and p are complex numbers. It is easy to see that both x and p, satisfy
Eqs. (7.7). You may like to check it out.
An Introduction to Quantum
Mechanics SAQ 2
Spend
5 min
Show that x and p, satisfy the criterion of linearity.

Secondly, in general, quantum mechanical operators do not necessarily commute. What

do we mean by this? To understand it, recall that in classical mechanics, we define
angular momentum L as r x p and not as p x r, which is equal to - L. Putting this
result in a mathematical language 'we can say that r and p do not commute under the
operation of vector product. A similar situation exists for quantum mechanical
operators. If two operators A and B operate one after the other on a function \y, then
their order of operation is important. In general, BAY is not equal to ABy for any
arbitrary \y, i.e.

The expression AB - BA is denoted by the commutator bracket [A, B]. Thus, we

-
define the commutator of two operators as the difference AB BA and denote it by the
symbol [A, B]:

and, in general,
r--1
[A, B] I AB - BA

[A, Bl 7t 0
In other words, in general, the operators A and B do not commute with one another and
the value of the commutation bracket [A, B] is non-zero. What does this result mean? It
means that we have to be careful,about the order of operators in considering operator
products in quantum mechanics. However, if the commutator of the operators A and B
vanishes, A and B commute, i.e., AB = BA. Then we can interchange their order.
To understand these concepts better, let us take a concrete example of operators. Let us
examine whether the operators x and p, commute with one another. For this purpose we
evaluate

Since \y is arbitrary we obtain

(7.10)
I

Thus we have found that x and p, operators do not commute with one another and the
value of the commutation bracket is iA, This result also tells us that we have to take
b
care of the order of these operators when we apply them on a system. For instance, if
the momentum operator acts first on a system followed by the position operator, it
yields a certain result. The result is different if the position operator operates first and is
followed by the momentumyoperator. This result has an interesting fallout. Sometimes,
you may come across a situation where the product of x and p occurs in a classical
aynamical variable. Now in quantum mechanics, the order of operators matters:So in
which order do & I put x and p? In such a case, we simply symmetrize the product,
-
i.e., we replace the variable xp, by the operator 1 (xp, + p p ) :
2

You should note that x and px are what are termed in classical mechanics as canonically
conjug&e 9ariables. In classical mechanics we do not have an equation like Eq. ( 7 . y ) ) .
since x and p, are dynarnical variables which have complex numerical values. So
they occur interchangeably in classical expressions of physical quantities. You 1 ,
should feel completely at home with these concepts before proceeding further. So work Observables and Operators
out this exercise.

--

SAQ 3 Spend
5 mitt
(a) ?how that xop commutes with (p,,,),p and (p,) ,p.
(b) Determine ly, p,,] and [z, p,l.

Thus you have found that the quantum mechanical operators corresponding to classical
canonically conjugate positio~iand momentum variables do not commute with one
another: x does not commute with p,, y does not commute with p), and z does not
colnmute with p,. The value of the commutation bracket is ifi in each case. Because of
this non-commutability we are required to write xp, ( y ~or, ~
zp,) in a symmetric form
while converting a dynamical variable D containing such terms into its quantum
mechanical operator. Let us now make use of the definition of [A, B] given by
Eq. (7.9b) to derive some interesting basic results of commutator algebra.

Basic Commutator Algebra

1. The following results satisfied by operators are useful and readily proved

[A, Bl = - [B, A1 (7.1 la)

[A, B + C ] = [ A , B] + [A, C ] (7.1 1b)

[AB, C ] = A [B, C ] + [A, CIB (7.1 1c)

and

[A, B C ] = B [A, C ] + [A, B]C (7.11d)

You should quickly verify Eqs. (7.1 la) to (7.1 Id) before studying further.
-1'
2. Any operator always commutes with its own power, i.e.,

It follows from Eqs. (7.11) and (7.12) that iff ( x ) is an operator which can be expanded
in the powers of x then

Similarly, iff (p,) can be expanded in the powers of p, we have

You may like to prove Eqs. (7.13) and (7.14) before studying further.

SAQ 4 Spend
10 mh
(a) Prove Eqs. (7.13) and (7.14).
(b) Show that the parity operator commutes with

Now, since observables are measurable attributes of any physical system, they are real
quantities. Therefore, they should be represented by operators which, when operating on
a physical system, yield real values of the observables. In quantum mechanics, all
An Introduction to Quaritum observables are represented by such operators, which are called hermitian operatom.
Mechanics Let us now study briefly about them.

Hermitian Operators
A hermitian operator is defined as follows:

For a one-dimensional system the volume element d2 is simply dx and the limit of
integration is from - m to + oa, However, for a three dimensional systems d2 is the
volume element drdy dz and all the three variables cover the whole space, i.e., the
variables vary from - m to + m.
Integrals like Eq. (7.15a) will occur quite often in tNs course. Hence we adopt a short
hand notation and take

and

I (0,4)* W = (D4,
yr)
Henceforth, we shall use the same symbol D for the dynamical variable and also for its
operator if there is no confusion. The integral (4, Dv)is also known as inner product
or scalar product of 4 with Dy. For $ = y and D = I, the identity operator, the
integral (yr, yr) is known as the norm of the wave function yr. You should.notice that
for a nomalised yr the norm is equal to unity. The norm of a wave function
representing a state of a system is always positive. It can be zero only when y = 0, i.e.,
that state of the system does not exist.
To understand these concepts concretely, let us now consider the linear momentum
operator p and show that it is a hermitian operator. We have p = f px + py + i? p,
't
where k and fi are unit vectors along x, y and r axes, respectively. You have already
proved in SAQ 2 that

This result holds for py and p, as well. Hence, px, py and p, are linear operators, Now
let us consider the integral

where pop stands for p,, py or p,. Integrating by parts we get

k
If at least one of the functions is a normalizable wave function then the first term
vanishes because the normalizable wave'functions go to zero at x = f 00. Thus

I=J;P,~)* Wd* (7.16)

4

implying that p, is a hermitian operator. Thus p,, py, p, and p are all hennitian
operators. . ,
The position operator x is obviously linear and hermitian. Hence the angular momentum
and Hamiltonian operators are also linear and hennitian,
The operators which satisfy E ~(7.15a)
. are also known as self adjoint operators. Here it
is useful to introduce the adjoipt or Hermitian conjugate of an operator Dl by the
relation.
If D = Dt then the operator D is said to be self adjoint. You can readily compare Ohmmblcs and Opmtorr
Eqs. (7.15a) and (7.17a) and see that for a hermitian operator
In quantum mechanics,
operators are represented
by matrices. In terms of
Now suppose D = AB then according to Eq. (7.17a) matrix algebra, the matrix
representing LIT is obtained by
taking the complex c~njagrrte
and the ttanspose of the
But we can also write matrix representing D.

Thus, applying Eq. (7.17a) twice we have

Hence, comparing Eqs. (7.18a) and (7.1 8b), we obtain an important result for adjoint
operatoe which applies to hennitian operators also:

So far, we have introduced you to the concepts of observables and operators. We have
said that every observable is associated with an operator. Now you may ask: Exactly
'what is the connection between obsehables and operators? That is what we shall
:discover in the next section.

7.2.2 ,Expectation Values

,Let us consider the measurement of a dynamical variable or the observable D of a
system. Keeping the system always in a particular state \y we measure D repeatedly. In
general, each individual measurement will yield a different result. Hence we take the
average of these measurements ( D ) as the value of the dynamical variable for that
particular state. Since we have always started with the same state y, it is reasonable to
assume that knowing \y we should be able to calculate (D). Such a relationship between
yr and (D) is provided by another postulate of the quantum mechanics. According to this
postulate,

The average of the measured value of D is given by Postulate 5:

(D) =
.b*DYfd - (yr. Dyr)
The meaeurcment
postulate
j W' WdT (w, '4')
(D) is known as the expectation value of the operator D.
, .l

If (D) is obtained from Eq. (7.19) comes out to be real then the dynamical variable D is
said to be an observable. Hence we can say that

An observable is a dynamical variable having

a real expectation vyue.

Now we can understand the significance of a hermitian operator. Hermitian operators

have real expsebtion values. To proye this result, we have
A n ~ c c l w t o where C is some constant representing .the no~alisationof \y. The complex conjugate
Mseblllrlcr of this equation gives

I
(D)* = C yr (Dy)* d~ (7.20b)
The difference is

Using Eq. (7.15a) this becomes

(D) - (D)* = C I@* (Dv) d~ - C jv*(Dv) dt

which means that (D) is real. Thus, we have proved that in quantum mechanics all
obsewubles are represented by hennitian operators.
We will now introduce you to another interesting feature arising out of the discussion so
far: Representing the Schriidinger equation as an eigenvalue-eigenfunction equation.

7.3 EIGENFUNCTIONS ANSD EIGENVALUES

So far you have studied that, in general, when an operator D operates upon yr we get a
new function yr'. However, under special circumstances yr' may just be a multiple of \ ~ r
itself, i;6.,

where d is a complex number. Under this situation, yr is said to be an eigenfunction of

the operator D having d as its eigenvalue. Eq. (7.22) is called the eigenvalue-
eigenfunction equation for the operator D. Now recall the time independent
Schriidinger equation given in Unit 6:

.
Do you recognise that the left hand side is nothing but the Hamiltonian H? We can also
write this equation as
Hyr=Eyr (7.23)
L

Thus, the time independent Schrtidinger equation is an eigenvalue eigenfunction

equation for the operator H (see Eq. 7.4). It tells us that H operating on a special class
of wave functions gives back the same wavefunctions multiplied by the eigenvalue E of
H. Since Hais hennitian, the eigenvalue E is real. This result can be proved for any
'
hennitian operator satisfying an eigenvalue eigenfunction equation as follows:
From E!q, (7.15a) we have

(yr, W )= (Dyr,yr)
Now with the help of Eq. (7.22) we get

But for a given state is not zero, hence (Y, yr) ia finite. Thus we obtain
d = d*

Hence, the eigenvdue of a hermitian operator is always real, In this case the ,
60 is equal to d itself, which is real.
expectation value (D)'
Using the concepts presented so far, we would like to introduce an important class of Observablm and Operators
eigenfunctions namely, eigenfunctions which are normalised to unity and satisfy the
orthogonality property. Such eigenfunctions are called orthonormal eigenfunctions. In
this connection, we will also introduce another useful concept of the degeneracy of
eigenfunctions.

Orthonormal Eigenfunctions
Suppose for a system there are more than one eigenfunctions of an operator having the
same eigenvalue. Then all such functions are called degenerate eigenfunctions.
Eigenfunctions of an operator having different eigenvalues are called non-degenerate
eigenfunctions. Let us now take two non-degenerate eigenfunctions Q and yr of a
hermitian operator D, having eigenvalues d l and d2, respectively:
D$ = dl$ and Dyr = d 2 v
Let these eigenfunctions be normalised to unity. Then from Eqs. (7.15a) and (7.22) we
obtain:

Since d l + d2 we find that the inner product (4, yf) of $ and y~ in Eq. (7.24) is zero.
Eigenfunctions having inner product equal to zero are said to be orthogonal to each
other:

We can generalise this statement. If vl,yr2, ... y~,,are non-degenerate eigenfunctions of

n hermitian oberator, nomalised to unity then they satisfy the following orthonormality
condition

(vi,v j ) = O for i + j (7.26a)

( ) = 1 for i = j (7.26b)
Thus, the eigenfunctions are normalised to unity, and all eigenfunctions yrl,vjsatisfy the
orthogonality property (7.26a) for i # j. We can make use of the Kronecker delta symbol
SI( and write Eqs. (7.26a and b) in a compact form:

where is defined as:

tiy = 0 for i sj
6,,= 1 for i = j

Such eigenfunctions which satisfy Eq.(7.26) are called orthonormal fbnctions and
form an orthonormal set.
Using these ideas we can show that if y is a non-degenerate eigenfunction of an
bperator D and D commutes with another operator B then y is also an eigenfunction of
B on Eq,(7.22) from the left to obtain
B. To pr'bve it let 'u~~operate

since d is a number. Furthermo$e, B commutes with D, hence, wa also have

 An introduction to Quantum - The above equation clearly shows that (Byr) is an eigenfunction of D with the same
Mcchanla eigenvalue d. Since y is not degenerate with BY, the eigenfunction B y must be a
multiple of y, i.e..

From Eq. (7.28) we conclude that y~ is also an eigenfunction of the operator B with the
eigenvalue b. In general, if there are n commuting operators and y is a non-degenerate
eigenfunction of any one of them then it is an eigenfunction of the remaining (n-1)
operators also. These n operators form a set of commuting operators.
We now proceed to demonstrate that if an operator A commutes with the parity
operator P, then the non-degenerate eigereunctions o f A have definite parity.

Let

Applying P to Eq. (7.29a) from left and ,using the condition [P, A] = 0 we get

) P y(x) are eigenfunctions of A with the same

Thus we notice that both ~ ( x and
eigenvalue. Since ~ ( xis) non-degenerate, the two functions y(x) and P y(x) can differ
at the most by a ccnstant. Hence

Thus v(x) is an eigenfunction of the parity operator with p as the eigenvalue. Applying
once again the parity operator we get

But v ( x ) and p2y (x) are identicnlly the same. Hence p2 = 1 , i.e., p = &I. Thus, ~ ( x )
are of definite parity. For degenerate eigenfunctions it is possible to take linear
) yr(-x) to obtain eigenfunctions of definite parity. Equation
combinations of ~ ( x and
(7.30b) also gives us the eigenvalues of the parity operator; these are 1. *
We shall use the parity operators in the next block to obtain eigenfunctions and.
eigenvalues of some simple systems,
We end this discussion with an exercise for you.

Spend SAQ 5
10 rnin
Show that the functions exp (-)?/2)and x exp ( - x 2 / 2 ) are eigellfunctions of the
operator (- d2/& -t 2).Calculate the eigenvalues and show that the two fullctions are
orthogonal to each other.

In the last section of this unit, it would not be out of place to establish a
' correspondence between the quantum mechanical and .Classical concepts. Remember that
in quantum mechanics we hqve operators and in classical mechanics there are only ,
dynamical variables which may be complex numbers. Thus we have to consider the
expectation values of operators. Now, according to the correspoqdence principle we
expect that the motion of a quantum object, represented by y,should agree with that of
a classical particle whenever the distances and momenta become so large that we can
ignore the uncertainty principle. When we try t0 explore this point, we arrive at the
Ehrenfest theorem,
I

62 I

i
 Ohservahles end Operators
7.4 EHRENPEST TMEOlREM
Let us consider the rate of change of an observable D, which does not depend explicitly
on time. From Eq. (7.19) we obtain for a normalized wave function y

-
d (D)
dt = (s. ~ v +) ( y ,D 3)
Now we use time dependent Schradinger equation in Eq. (7.31) to replace ayr by
-
1 at
i
, -
iA
Hv. Thus

=d ( D ) - I [ jy* DH yr a% -jyPHD r dt], since H is hermitian

dt ih
H+ = H.

Let us now take D to be ,position operator then

Putting Eq. (7.33) into Eq. (7.32b) we obtain

'
Furthermore, let us take D to be the linear momentum operator, In this case

You should note that Eqs. (7.34) and (7.35) are very similar to the equations which
define linear momentum and force in classical mechanics. However, the basic difference
between the two mechanics is that J, p, and 2 of classical mechanics a n repluced by
ax
their average values in quantum mechanics. For a macroscopic system there is hardly
any difference between x, p, and av/ax and their average values. However, for
microscopic systems they are quite different. As a matter of fact you have seen that for
a microscopic system the precise values of x and p, do not exist simultaneously but
their average values (x) and ( p , ) can be obtained.
Eqs. (7.34) and (7.35) constitute the Ehrenfest theorem which shows the
correspondence as well as a basic difference between classical and quantum mechanics,
You may like to apply these ideas and make use of Eq. (7332b) to arrjve at an
interesting result.
An Intraductlbn to Quantum
Mechanlce
SAQ 6
Spend
5 min Show that when an operator commutes with the Hamiltonian, the expectation value of
the observable associated with it is a constant of motion. Hence prove that the linear .
momentum of a system is conserved when no net force fields acts on the system.

Let us now summarise what you have studied in this unit.

7.5 SUMMARY
8 In this unit you have learned about two more postulates of the quantum mechanics,.
in addition to the postulates given in the previous unit. According to the first of
these postulates, every observable is associated with an operator. The operators
corresponding to the dynamical variables x and p, are x and -2 -
3.
ax
e To construct an operator of any other dynamical variable we write that function in
terms of x, p, (in a symmetric form) and then replace p, by - i A alax.
@ Most of the operators relevant to quantum mechanics are linear and hermitian, i.e.,

@ According to another postulate of quantum mechanics the expectation value of a

dynarnical variable Dl is equal to the average value of D, obtained by the ,repeated
measurement of D for that system in the same state. A dynamical variable having
real expectation value is said to be an observable.

8 For quantum mechanical operators in general ABy # BAyl and the value of the
commutation bracket [A, B] = AB - BA is non-zero.
e If th'e operation of D on y~ produces a multiple of y~ say @ then yl is said to be an
eigenfunction of D having eigenvalue d . The eigenvalues of a hermitian operator
are real.
0 The rate of change of average (x) and ( p , ) for a system of mass m and potential
energy V(x) are equal to ( p , ) / m and (- aV/ax), respectively. These relations are
called Ehrenfest theorems and are very similar to those obtained in classical
mechanics with the difference that in classical mechanics we consider x, p, and
aVlax themselves instead of their averages, ,

7.6 TERMINAL QUESTIONS Spend 45 rnin

1. A state of a particle of mass nt is given by e-&. Normalise the wave function

.and calculate the expectation value of the kinetic energy of the particle.
2. If for two operators A and B

[A, B] = 1
then show that [Az, B2] = 2 (AB + BA).

3. If two non-commuting operators A and B commute with their commutator [A, Bl,
show that
[A, B"] = n B"-' [A, B]
where n is an integer. Hence obtain the value of [B, p,],
 4. If for a quantum mechanical system Obeervables and Operatom

(p,2/2m + V(x)) v ( x ) = E v ( x )
(K.E.) = - 1 (X av/ax>.
show that
2
The expression is known as Virial theorem.
Hint: Start with ([xp,, H 1 ) = 0.

5. a) Determine whether the parity operator P is hermitian or not.

b) Show that all operators which are invariant under space inversion commute with
the parity operator.
6. The Hamiltonian of a system is given by
-2

and yrl(x) and v z ( x ) are two degenerate energy eigenfunctions of the system. Show
that

,
, x21 V2) = 0.
Hint: Start with (WI,[H,

7. Show at [L,, L,] = iA L,,. Hence prove that if 4 is an eigenfunction of L, ,

't(LSY>= (Lx) = 0,
8. Suppose v ( x ) = Ci ci @i (x)
where @ i ( ~ )are eigenfunctions of a hermitian operator D with eigenvalues d, and
(4i, $j) = 6$ then show that

(w, Dv) = C di I ci I2

7.7 SOLUTIONS AND ANSWERS

Self-Assessment Questions

= - i A z - -a+ iAx- a
ax az
Lz = XP,Y - YPx
AII raboduetlon to Quantum
Mechaab

= a P,Y + b Px 4
Therefore, x and p, are linear.

= - i f t ~ ra'-! + il) x-
JY (': x and y are independent)
ay ay
=0

Since yr is arbitrary,

"P), - pyx = 0
Thus operators x and py commute.
d

Similarly, we can show that x,commbtes with p,.

[z,pz1 = il)
Proof is similar to that of [x, px] = iA
4. (a) Since f (x) can be expanded in powers of x, we may write

Now using Eq. (7.1 1c) we can write

and [J?, pxl = X [ 2 , PA + [x, PXP '

= x [2 IAx] + iAx2
= 31'lt2
Similarly,

Thus,we have
. .

= ilt [l + 2r + ?.3 + .., + me',+ ...I ,

< = iA
.a
=&f(x)

You can prove Eq. (7,141 in the same way.

since A (-XI = 4x)"hence PA (x) y (x) = A (-x) yr (-x) = A(x) Py (x). Since yr (x) is
arbitrary, A(x) commutes with P.

-
 5, We have to show that Observables and Operetom, .
I

and calculate h where

and

= eeZn implying hl = 1.

Terminal Questions
DD

1. Tho no-liaation condition is N

' dx = 1
r2"$

{K.E.) = {p2/2m) = (2a/%)In (el, *

- 5 a2,2,

1
or {K.E) = y u #/z/m.
2. [A Z, $1 = A [A, @I + [A, B2lA (Using Eq. 7.11~)
. = A {B[A, 81 + [A, 81 B ) + {B[A, B] + [A, B] B ) A

3. Let [A, Bn] = n B"-'[A, B] (1)

Hence from Eq. (7.l Id)
[A, @"+'I = B[A, @I] + [A, B] BN
 = n B [A, B] + [A, B] Bn (Using Eq. 7.14)
= (n + 1) B" [A, B]. (2)
Hence if ( 1 ) is true for n it is also true for n + I. Since (1) is certainly true for n = 1
hence it is also true for n = 2. Thus Eq. (1) is true for any n.

Now [xpxl HI = bp,, p:/2m + V(4l

~ 7.1 Id)
From Eqs. (7.1lb, 7 . 1 1 and

:a Using (1) we get (K.E)= 1/2 ( x a ~ / a x ) .

5. a) We have
DO DO ' .-oo

j W*(X) p yr(4 dr = j w*(x) v ( - x ) h = - j v (-x

--00 m I
7 y ( x ')dr '
where x '= -x
= J;w
-00
(*')I* v ( x 7 dx'

Hence P is a Hermitian operator.

b) PA(x, P,) ~ ( x=)A(+, -P,) v(-x) = A (-xl -pX) P V(X)= A (x,P,)P W(x)
Hence (PA (x, P,) - A(xl P,)P) yr (4= 0
Since yt (x) is. arbitrary we have

[PIA] = 0,
6. Since yr1 and yrz are degenerate, we have

1 = [ p,2/2rn + V(x),A?]
Now [HI2 .

= ( l / m ) (p,x-.+ xp,] (Using the result of terminal question 2)

.'a (ytll ( P 9 + xe,) yrz) = 0. I

. 68 '
 7. [L,, L,l = L&, - LAz Observables and Operatars

Now from the definition of Lx and L, (see SAQ l(b)) we have

L, = yp, - zp), and L, = xp), - YPX

Hence [L,, L,I = [xPY- YPX,YPz - zP,~l

Using Eq. (7.1 1b) we get
[L,, L,] = [xpY, YP,] - [XP~!,
zpJtl - [YP, YP,] + [ Y P ZP~J
~
Using Eqs. ( 7 . 1 1 ~and d)
[L,, Lxl = y [xp,,, P,] + [xP,!, YIP,- 0 - 0 + Y [p, zfil + [Y, zpy1px
= 0 - ih xp, + iA zp,
= itt (zpx - xpz) = iA Ly

Let t,$ = m 4
But LA, - L.4, = ifi L.v

:. (4, L, Lx4) - (4, L&,4) = iR (Ly)

m (Lx) -m (L,) =. ift (L), )
:. (L.v)=o:
Similarly, (L,) = 0.

8. Since V(X) = Cci $i(x)

i

we have (D) = C ZC,*cj 4 (&. $j)

i
= Zc;cj d1 6V
I' j

= C I c, I= 4
i

since only those terms of the j series will survive for which j = i:

. FURTHER READING
I , Concepts of Modern Physics, A. Beiser, McGraw-Hill International ~ o o k - ' ~ o r n ~ a n ~ ,
1990.
2. Introduction to Quantum Mechanics, B.H. Bransden, C.J. Joachain, ELBS, 1990.
3. Quantum ~ e c h a n i c s J.L.
, Powell, B. Crasemann, Addison Wesley Inc, 1961.
4. A Textbook of Quantum Mechanics, P.M. Mathews, K.Venkatesan, Tata McGraw
Hill Publishing Company Limited, 1987.
A" Introduction to Quantum
Mcchanics A PERSPECTIVE ON QUANTUM MECHANICS
In this block we have introduced you to those basic ideas and concepts which form the
bulwark of the new quantum mechanics. In the process, it may have seemed to you that
the entire edifice of classical physics has been turned upside down: the classical ideas of
causal determinism, continuity, unambiguous and precise language descriptions lie
squarely challenged. What has replaced it is an entirely new way of thinking and
understanding our world. And because the behaviour of objects in the quantum world is
so unlike ordinary experience, you nlay have found it very difficult to get used to it, in
the first instance. Do not worry. It appears peculiar and mysterious to everyone who
encounters it for the first time - whether a novice or an experienced physicist.

All of us know how large objects act - all of our direct experience and intuition
applies to such objects. But as you have studied in this block, things on n small scale
just do not act that way. Quantum objects are wave-patticles represented by wave
functions. Though the time-evolution of wave functions is govermed by an equation of
motion, its solutions give us only a probability of finding the wave-parlicles in a certain
region at a given time. Measurement of the physical observables like position, momenta,
energy, etc. (which call be determined precisely for classical objects) is governed by the
uncertainty principle in the quantum mechanical world. Then there is the idea of
quantum jumps (or discontinuities) in quantum mechanics. To put it in a nutshell, as per
the Copenhagen interpretation of quantum nlechanics cleveloped tluough the ideas of
Born, Heisenberg and Bohr, we calculate quantum objects probabilistically, we
determine their attributes somewhat uncertainly and we understand them
complementarily. Quantum mcchanics, thus, presents a new and exciting world-view that
challenges old concepts such as deterministic trajectories of motion and causal
continuity. It springs unexpected surprises on us, and keeps our minds in a constallt
flurry of animated activity.

For those of you philosophically inclined, we present here an excerpt from Feynman's
Lectures on Physics which gives us a perspective on quantum mechanics. It is a
masterly reflection upon one of [he most fundamental concepts of quantum mechanics -
the uncertainty principle which has unendingly troubled the best of minds. Through
Feynman's eyes, we, the students of physics, get to look deeply and philosophically into
the nature of quantum mechanics and the nature of science. This, in our opinion,
constitutes a befitting final6 to an introductory foray into the world of quantum
mechanics.

"Philosophical Implications

Let us consider briefly some philosophical implications of quantum mechanics. As

always, there are two aspects of the problem; one is the philosobhical implication for
physics, and the other is the extrapolation of philosophical matters to other fields. When
philosophical ideas associated with science are dragged into another field, they are
usually completely distorted. Therefore we shall confine our remarks as much as
possible to physics itself.

First of all, the most interesting aspect is the idea of the uncertainty principle; making
an observation affects a phenomenon. It has always been known that making
observations affects a phenomenon, but the point is that the effect cannot be disregarded
or minimized or decreased arbitrarily by rearranging the apparatus. When we look for a
~ertaii'~henomenon we cannot help but disturb it in a certain minimum way, and the
disturbartce is necessury for the consistency of the viewpoint. The, observer was
sometimes important in prequantum physics, but only in a rather trivial sense. The
problem has been raised: if a tree falls in a forest and there is nobody there to hear it,
does it make a noise? A real tree falling in a real iorest makes a sound, of course, even
if nobody is there. Even if no one is present to hear it, there are other traces left. The
sound will shake some leaves, and if wc were careful enough we might find somewhere
that some thorn had rubbed against a leaf and made a tiny scratch that could not be
explained ullless we assumed the leaf were vibrating. So in a certain sense we would
have to admit that there is sound made. We might ask; was there a serisation of sound?
No, sensations have to do, presumably, with consciousness. And whether ants are
 conscious and whether there were ants in the forest, or whether the tree was conscious,
we do not know. Let us leave the problem in that form.
Another thing that people have emphasized since quantum mechanics was developed is In science we go by experiments
the idea that we should not speak about those things which we cannot measure. - even conceptual experiments.
(Actually relativity theory also said this.) Unless a thing can be defined by measurement, And if the limitation is not of the
nctunl measuring devices used hut
it has no place in a theory. And since an accurate value of the momentum of a localized is set by the fundamental processes
particle cannot be defined by measurement it therefore has no place in the theory. The of measurement then we have to
idea that this is what was the matter with classical theory is a fal~eposition. It is a accept it.
careless analysis of the situation. Just because we cannot measure position and
momentum precisely does not a priori mean that we cannot talk about them. It only
means that we need not talk about them. The situation in the sciences is this: A concept
or an idea which cannot be measured or cannot be referred directly to experiment may
or may not be useful. It need not exist in a theory. In other words, suppose we compare
the classical theory of the world with the quantum theory of the world, 'and suppose that
it is true experimentally that we can measure position and momentum only imprecisely.
The question is whether the ideas of the exact position of a particle and the exact
momentum of a particle are valid o~ not. The classioal theory admits the ideas: the
quantum theory does not. This does not in itself mean that classical physics is wrong.
When the new quantum mechanics was discovered, the classical people-which included
everybody except Heisenberg, Schrodinger, and Born-said: "Look, your theory is not
any good because you cannot answer certain questions like: what is the exact position of
a particle?, which hole does it go through?, and some others." Heinsenberg's answer
was: "I do not need to ask such questions because you cannot ask such a question
experimentally." It is that we do not have to. Consider two theories (a) and (b): (a)
contains an idea that cannot be checked directly but which is used in ,the analysis, and
the other, (b) does not contain the idea. If they disagree in their predictions, one could
not claim that (b) is false because it cannot explain this idea that is in (a) because that
idea is one of the things that cannot be checked directly. It is always good to know
which ideas cannot be checked directly, but it is not necessary to remove them all. It is
not true that we can pursue science completely by using only those concepts which we
directly subject to experiment. Let us give you an idea about the
debate on determinism (cnusnlity)
In quantum mechanics itself there is a wave function amplitude, there is a potential, and versus free will: According to
there are many constructs that we cannot measure directly. The basis of a science is its Newtoninn dynamics, if the
position, velocity md the forces
ability to predict. To predict means to tell what will happen in an experiment that has niting on n body at any instnnt of
never been done. How can we do that? By assuming that we know what is there, time are known its 'state' nt all
independent of the experiment. We must extrapolate the experiments to a region where Inter times can be p~dicted
, they have not been done. We must take out concepts and extend them to places where accurnkly. Thnt is, if know the
'cause' we can predict the effect.
they have not yet been checked. If we do not do that, we have no prediction. So it was This applies to any object, however
perfectly sensible for the classical physicists to go happily along and suppose that the large or small. Extending this to nll
position-which obviously means something for a baseball-meant something also for objects in the universe it wm
an electron. It was not stupidity. It was a sensible procedure. Today we say that the law thought thnt every event and ih
time evolution can be determined
of rgatlvity is supposed to be true at all energies, but somebody may come along and for nll times to come. This
say how itupid we were. We do not know where we are "stupid" until we "stick our determinism would apply even to
,neck out;" and so the whole idea is to put our neck out. And the only way to find out humnn body nnd mind. This means
that we are wrong is to find out what our predictions are. It is absolutely necessary to that even the humnn mind has no
free will, no freedom of choice;
make constructs. the future of eyery living ckture,
being n pnrt of the mechmistic
We have already made a few remarks about the indeterminacy of quantum mechanics. universe is completely determined.
That is, that we are unable now to @edict what will happen in physics in a given
physical circumstance which is arranged as cmefully as possible. If we have an atom Now things ore entirely different In
qunntum mechanics. We can only
that is in an excited state and so is going to emit a photon, we cannot say when it will predict the probability of nn event
emit the photon. It has a certain amplitude to emit the photon at any time, and we can taking plnce nnd the nttributes of a
' predict only a probability for emission; we cannot predict the future exactly. This has system me governed by the
given rise to all kind of nonsense and questions on the meaning of freedom of will, and uncertainty principle. Thus, there
seems to k n complete brenkdown
of the ideas that the world is uncertain. of determinism (or the cause-effect
relationship in the rnicroscoplc
j Of course we must emphasise that classical physics is also indeterminate, in a sense. It world). This is interpreted by soma
is usually thought that this indeterminacy, that we cannot predict the future, is a n . as restomtion of 'freedom of will'
'important quantum-mechanical thing, and this is said to explain the behaviour of the to choose an nlternntive in the
mind, feelings of free will, e t c . . ~ u tif the world were classical-if the laws of probabilistic world which is also
uncertnin..
mechanics were classical-it is not quite obvious that the mind would not feel more or
less the same.-It is true classically that if we knew the position and the velocity of
 -
An I&oduetlon to ~ u l ~ t u r n every particle in the world, or in a box of gas. we could predict exactly what would
Mechdea, happen. And therefore the classical world is deterministic. Suppose, however, that we
have a finite accuracy and do not know exactly where just one atom is, say to one part
in a billion. Then as it goes along-it hits another atom, and because we did not know
the position better than to one part in a billion, we find an even larger error in the
position after the collision. And that is amplified, of course, in the next collision, so that
if we start with only a tiny error it rapidly magnifies to a very great uncertainty. To
give an example: if water falls over a dam, it splashes. If we stand nearby, every now
and then a drop will land on our nose. This appears to be completely random, yet such
a behavior would be predicted by purely classical laws. The exact position of all the
drops depends upon the precise wigglings of the water before it goes over the dam.
How? The tiniest irregularities are magnified in falling, so that we get complete
randomness. Obviously, we cannot really predict the position of the drops unless we
know the motion of the water absolutely exactly.
Speaking more precisely, given an arbitrary accuracy, no matter how precise, one can
find a time long ebough that we cannot make predictions valid for that long a time.
Now the point is that this length of time is not very large. It is not that the time is
millions of years if the accuracy is one part in a billion. The time goes, in fact, only
logarithmically with the error, and it turns out that in only a very, very tiny time we
lose all our information. If the accuracy is taken-to be one part in billions and billions
and billions-no matter how many billions we wish, provided we do stop somewhere-
then we can find a time less than the time it took to state the accuracy-after which we

/
can no longer predict what is going to happen! It is therefore not fair to ay that from
the apparent freedom and indeterminancy of the human mind, we shou have realized
that classical "deterministic" physics could not even hope to understa d it, and to
welcome quantum mechanics as a release from a "completely mechanistic" universe. For
already in classical mechanics there was indeterminability from a practical point of
view."
 Table of fundamental constants

\ .
Ouantitv Symbol Value

Planck's constant

Velocity of light in vacuum

Elementary charge (absolute
value of electron charge)
Permeability of free space

~erinh.ivityof free space

Gravitational constant
Fine structure constant

Avogadro's number
Faraday's constant
Boltzmann's constant

Gas constant

Atomic mass unit

Electron mass

Proton mass

Neutr~n,mass

Ratio of protbn to MpL'

electron mass
~\actroncharge to
mass ratio -
Classical radius
'of electron
~XE~A*
3ohr radius for a. =
'

- 2
5,29177 X 10-I m *
atomic hydrogen me
\, (with infinite nuclear mass) ',

2
a
Rydberg'a constant R ~ = - =me- - 1.09737 x lo7 m-'
for infinite nuclear mass 8c2,,h3c 4m0

I Rydberg's constant
for atomic hydrogen
eft
Bohr rnagneton 9.27408 x J T1
P~=-%T
Nuclear magneton pN= eA . - 5.05082 x 10-" JT1
. 2MP
I . .

t .. ..