Download as pdf or txt
Download as pdf or txt
You are on page 1of 69

Oscillations and Waves

PH-1007 (Physics)
Dr. Gorky Shaw

1.1 Simple harmonic motion


1.1.1 Introduction
1. Periodic motion
Repeated motion along a definite path at regular intervals of time is called
periodic motion. The interval of time is called time interval.

2. Oscillation or vibratory motion


To and fro or back and forth motion repeatedly about a mean position is
called oscillation or vibratory motion.
Such periodic and bounded motion is confined within well-defined limits
on either side of the mean position. The body performing oscillation is
called oscillator.
To represent this type of motion mathematically, sine and cosine functions
are best suited, being both simple periodic and bounded.

3. Simple harmonic motion (SHM)


Oscillatory motion is in which the amplitudes are equal on both sides of
the mean position is called simple harmonic motion (SHM).
Examples of SHM are motion of a spring and simple pendulum.

Figure 1.1: Graphical representation of simple harmonic motion.


1.1 Simple harmonic motion
4. Necessary condition for SHM
A necessary condition for SHM is a restoring force acting on the body.

F ∝ −y
or, F = −ky (1.1)

where y is the displacement of the body from its mean position. k is


called force constant or spring constant.

1.1.2 Equation of motion and solution


1. Differential equation for SHM.
For any body of mass m, being acted upon by a force F and having
velocity v and acceleration a, the equation of motion is,

F = ma (1.2)

Now,

dv d2 y
a= = 2 (1.3)
dt dt

Therefore, from (1.1) and (1.2) and (1.3), we get

d2 y
m 2 = −ky
dt
d2 y k
=⇒ 2 + y = 0
dt m
2
dy
=⇒ 2 + ω 2 y = 0 (1.4)
dt

k
where ω 2 = is a positive constant.
m

(1.4) is the differential equation representing SHM.

A special model for SHM, the simple pendulum, is discussed later in


Section 1.6.1.
The next step is to find a solution of this equation to characterize the
motion.

2
1.1 Simple harmonic motion

2. Solution of the equation and expression for the displacement in


SHM
By making an educated guess, we can see that two possible solutions of
(1.4) are

y = eiωt and y = e−iωt (1.5)

which we call y1 and y2 , respectively.


In such cases, it is known that, since y1 and y2 are solutions of (1.4), their
linear combination

y = C 1 y 1 + C 2 y2
or, y = C1 eiωt + C2 e−iωt (1.6)

where C1 and C2 are constants (usually complex numbers) to be deter-


mined from initial conditions, is also a solution of (1.4).

Exercise: Substitute y1 = eiωt , y2 = e−iωt , and y = C1 y1 + C2 y2 in (1.4)


and check that they all are indeed solutions of this equation.

Now, we know that

eiωt = cos ωt + i sin ωt


and, e−iωt = cos ωt − i sin ωt (1.7)

Substituting (1.7) in (1.6), we get

y = C1 (cos ωt + i sin ωt) + C2 (cos ωt − i sin ωt)


=⇒ y = (C1 + C2 ) cos ωt + i(C1 − C2 ) sin ωt (1.8)

We make the replacement of variables:

A1 = C1 + C2
and, A2 = i(C1 − C2 ) (1.9)

Thus, from (1.6), (1.8), and (1.9), General solution of (1.4) is,

y = A1 cos ωt + A2 sin ωt (1.10)

Where A1 and A2 are constants to be determined from initial conditions.

3
1.1 Simple harmonic motion

For example, we consider the following initial condition: at t = 0, y = 0.


Putting this in (1.10), we get

0 = A1 + 0 =⇒ A1 = 0 (1.11)

y = A2 sin ωt for this initial condition. (1.12)

Exercise: Check that y given in (1.12) satisfies (1.4).

We introduce new variables A and φ, related to A1 and A2 as:

A1 = A sin φ
and, A2 = A cos φ (1.13)

Then, substituting (1.13) in (1.10), we get

y = A sin φ cos ωt + A cos φ sin ωt


or, y = A(sin φ cos ωt + cos φ sin ωt)
or, y = A sin(ωt + φ) (1.14)

Hence, we may rewrite the general solution in a more convenient form,

y = A sin(ωt + φ) (1.15)

where the term (ωt + φ) is called the phase of the SHM, and φ is called
the initial phase or phase constant.

Using the above initial condition: at t = 0, y = 0 in (1.15) we get,

0 = A sin φ =⇒ φ = nπ, n = 0, 1, 2, ... (1.16)

Without loss of generality we can take the value φ = 0.

Therefore, y = A sin ωt for this initial condition. (1.17)

(1.6), (1.10), and (1.15) are all valid forms of the general solutions of
(1.4). However, (1.15) is the most convenient form to represent and study
simple harmonic motion.

4
1.1 Simple harmonic motion

3. Parameters associated with SHM


The parameters associated with SHM are:
ˆ A → Amplitude of oscillation (maximum displacement from the
mean position).
ˆ ω → angular frequency.
ˆ f → frequency. ω = 2πf
1 2π
ˆ T → Time period of oscillation. T = = .
f ω
4. Expressions for velocity and acceleration associated with SHM
Using the expression for y in (1.15) the velocity v is given by,
dy
v= = Aω cos(ωt + φ) (1.18)
dt q
= ±Aω 1 − sin2 (ωt + φ)
q
= ±ω A2 − A2 sin2 (ωt + φ)
p
= ±ω A2 − y 2 (1.19)

Velocity is represented by either (1.18) or (1.19).


Acceleration a is given by,

d2 y
a = 2 = Aω 2 sin(ωt + φ) (1.20)
dt
= −ω 2 y (1.21)

Acceleration is represented by either (1.20) or (1.21).

1.1.3 Expression for energy in SHM


In order to determine the energy of the oscillator, we first evaluate its kinetic
energy and potential energy separately and then add them to get the total
energy.
Kinetic energy of the oscillator,
1 1
K = mv 2 = mω 2 (A2 − y 2 ) (using (1.19)) (1.22)
2 2
Clearly, Maximum and minimum values of kinetic energy are at the mean
position and extreme position, respectively.

5
1.1 Simple harmonic motion

1
Kmax = mω 2 A2 at y = 0
2
Kmin = 0 at y = A (1.23)

Potential energy at displacement y is the work done against the restoring


force in moving the body from the mean position to this position:
Z y Z y
P = −F dy = kydy
0 0
1
= ky 2
2
1
= mω 2 y 2 (since k = mω 2 ) (1.24)
2
Clearly, Maximum and minimum values of potential energy are at the extreme
position and mean position, respectively.

1
Pmax = mω 2 A2 at y = A
2
Pmin = 0 at y = 0 (1.25)

The total (mechanical) energy,


1
E = K + P = mω 2 A2 (1.26)
2
is conserved (constant).

6
1.2 Damped Harmonic Oscillations

1.2 Damped Harmonic Oscillations


As discussed in Section 1.1, the amplitude of oscillations remains constant in
case of simple harmonic oscillations (SHM).
When the amplitude of oscillations of an oscillator reduces due to an external
force, the oscillator and its motion are said to be damped.

1.2.1 Equation of motion


The forces acting on such an oscillator are:

1. Restoring force:

FR ∝ −y
or, FR = −ky (1.27)

2. Damping force: A retarding (decelerating) force that is proportional to


its velocity v.
dy
FD ∝ −
dt
dy
or, FD = −b (1.28)
dt

Hence, the equation of motion for a oscillator of mass m is

d2 y
m = F R + FD
dt2
d2 y dy
or, m 2 = −ky − b
dt dt
d2 y dy
=⇒ m 2 +b + ky = 0
dt dt
d2 y b dy k
=⇒ + + y=0 (1.29)
dt2 m dt m
b k
Let = 2r and = ω02 . Then we get
m m

d2 y dy
2
+ 2r + ω02 y = 0 (1.30)
dt dt
This is the equation of motion of a damped harmonic oscillator.
r is called the damping constant, and ω0 is called the natural frequency of the

7
1.2 Damped Harmonic Oscillations

oscillator. Natural frequency is the frequency of the oscillator in the absence of


any damping (i.e., for r = 0).

1.2.2 Solution of the equation and expression for the displacement


In Section 1.1.2, we noticed that the solution of the equation for SHM involves
exponentials. Similarly, for the case of damped motion, we consider a possible
solution of the form

y = eαt (1.31)

where α is a parameter of motion and is to be determined.


(Note that in case of SHM we directly guessed the solutions y1 = eiωt and
y2 = e−iωt , which worked. However, in the case of damped motion, we consider
a more general exponent α which might depend both on r and ω0 .)
Substituting (1.31) in (1.30), we get

(α2 + 2rα + ω02 )eαt = 0 (1.32)

Thus, we get

α2 + 2rα + ω02 = 0
q
=⇒ α = −r ± r2 − ω02 (1.33)

Hence, we have two possible values of α:


q
α1 = −r + r2 − ω02
q
and α2 = −r − r2 − ω02 (1.34)

and hence two possible expressions for y:



r2 −ω02 )t
y1 = e(−r+

(−r− r2 −ω02 )t
and y2 = e (1.35)

8
1.2 Damped Harmonic Oscillations

Hence, the general solution of (1.30) is given by

y = A1 y1 + A2 y2
√ √
(−r+ r2 −ω02 )t (−r− r2 −ω02 )t
or y = A1 e + A2 e (1.36)

where A1 and A2 are constants of motion, to be determined from initial condi-


tions.

Based on the extent of damping, there are three distinct cases of damped
harmonic motion:
1. r2 < ω02 : underdamped motion.
2. r2 > ω02 : overdamped motion.
3. r2 = ω02 : critically damped motion.
As we will see subsequently, in the different cases, the displacement varies
with time as represented graphically in Figure 1.2.

Figure 1.2: Displacement versus time in damped motion (Source: math24.net).

1. Underdamped motion (or light damping)


In this case, r2 < ω02 , which means (r2 − ω02 ) < 0. We introduce a new
parameter ω given by
q
2
q
2
√ q 2
r2 − ω0 = −(ω0 − r2 ) = −1 ω0 − r2 = iω (1.37)

9
1.2 Damped Harmonic Oscillations
p
where ω = ω02 − r2 is a real number and represent the angular frequency
of damped oscillations. Clearly, ω < ω0 .
Hence, we can rewrite (1.36) as

y = A1 e(−r+iω)t + A2 e(−r−iω)t
y = A1 e−rt eiωt + A2 e−rt e−iωt
= e−rt (A1 eiωt + A2 e−iωt )
= e−rt [A1 (cos ωt + i sin ωt) + A2 (cos ωt − i sin ωt)]
= e−rt [(A1 + A2 ) cos ωt + i(A1 − A2 ) sin ωt] (1.38)

Now, since we are discussing oscillations, which are very real, the displace-
ment y is certainly a real quantity. Hence, both (A1 + A2 ) and i(A1 − A2 )
must be real quantities.
Accordingly, we make the replacement of variables:

A1 + A2 = A0 sin φ
and i(A1 − A2 ) = A0 cos φ (1.39)

Then, from (1.38) and (1.39), we have the general solution for under-
damped harmonic oscillations

y = A0 e−rt sin(ωt + φ) (1.40)

Figure 1.3: Graphical representation of underdamped oscillations.

10
1.2 Damped Harmonic Oscillations

2π 2π
which represents harmonic oscillations with time period T = =p 2
ω ω0 − r 2
As mentioned earlier, ω0 is the natural frequency (without any damping)
and ω is the damped frequency.
The amplitude of oscillations

A = A0 e−rt (1.41)

decays exponentially with time.

2. Overdamped motion (or hard damping)


p
In this case r2 > ω02 . Hence, r2 − ω02 is a real number.
p
Let r2 − ω02 = β.

Then for the overdamped case, the general solution given by (1.36) can
be rewritten as

y = A1 e(−r+β)t + A2 e(−r−β)t (1.42)

Now clearly, r > β.


Therefore, both the exponential terms in (1.42) decay with time. Thus,
in this case there are no oscillatory terms, and the resultant motion is
non-oscillatory. This type of motion is also called aperiodic motion.

3. Critically damped motion


Critical damping occurs when the damping constant is equal to the nat-
ural frequency of the oscillator. In this case r = ω0 .
Then the general solution given by (1.36) is reduced to

y = A1 e−rt + A2 e−rt = (A1 + A2 )e−rt = Ce−rt (1.43)

where we introduce a new parameter C = A1 + A2 , which is a single con-


stant term in the expression of y. Hence, (1.43) is insufficient as a general
solution of a second order equation (which must include two distinct con-
stants, as seen in all the earlier general solutions).
p
Therefore, in this case, we instead consider r2 − ω02 = h → 0 but 6= 0.

11
1.3 Energy decay in underdamped harmonic oscillations

Then, we have

y = A1 e(−r+h)t + A2 e(−r−h)t
= e−rt (A1 eht + A2 e−ht )
= e−rt [A1 (1 + ht + ...) + A2 (1 − ht + ...)]
= e−rt [(A1 + A2 ) + h(A1 − A2 )t] (1.44)

In the above expression, we have replaced the exponentials involving h in


the power with the respective series expansions, and ignored the higher
order terms since h is very small.
We further rewrite (1.44) as

y = e−rt (P + Qt) (1.45)

which is the general solution for critically damped motion.

where P = A1 + A2 and Q = h(A1 − A2 ). We can find P and Q from


initial conditions.
Initially, y given by (1.45) increases due to the linear term in t. But
with increasing t, the exponential decay term dominates, and y decreases
rapidly.

Critical damping provides the quickest approach to zero amplitude for a


damped oscillator. With overdamping, the approach to zero is slower. In
case of underdamping, zero is reached more quickly, but is crossed, and there
is oscillation about zero.

1.3 Energy decay in underdamped harmonic oscillations


For underdamped oscillations, we have (from (1.40))

y = A0 e−rt sin(ωt + φ) (1.46)

Therefore,
dy
v= = ωA0 e−rt cos(ωt + φ) − rA0 e−rt sin(ωt + φ)
dt
= A0 e−rt [ω cos(ωt + φ) − r sin(ωt + φ)] (1.47)

Now, for small damping, r << ω.

12
1.4 Parameters associated with underdamped harmonic oscillations

Therefore, in this case,


dy
v= ≈ A0 e−rt [ω cos(ωt + φ)]
dt
= ωA0 e−rt cos(ωt + φ) (1.48)

Therefore, kinetic energy of the oscillator,


1 1
K = mv 2 = mω 2 A20 e−2rt cos2 (ωt + φ) (1.49)
2 2
and its potential energy
1 1
P = ky 2 = mω 2 y 2
2 2
1
= mω 2 A20 e−2rt sin2 (ωt + φ) (1.50)
2
Therefore, total energy
1
E = K + P = mω 2 A20 e−2rt [cos2 (ωt + φ) + sin2 (ωt + φ)]
2
1
= mω 2 A20 e−2rt (1.51)
2
Hence, we may write

E = E0 e−2rt (1.52)

1
where E0 = mω 2 A20 us the energy at t = 0.
2
Thus, the energy of a damped harmonic oscillator decays exponentially with
time.
Note that while amplitude decay is proportional to e−rt , energy decay is
proportional to e−2rt . This is expected because the energy of oscillation is
proportional to the square of the amplitude, as seen in the above relations.

1.4 Parameters associated with underdamped harmonic


oscillations
1. Logarithmic decrement (λ): it is ratio between successive amplitudes,
at time intervals of T /2.
The displacement for underdamped oscillations is given by (1.40):

y = A0 e−rt sin(ωt + φ) = A sin(ωt + φ) (1.53)

13
1.4 Parameters associated with underdamped harmonic oscillations

where A0 initial amplitude at t = 0, and A = A0 e−rt is the amplitude at


time t.

Figure 1.4: Displacement versus time for underdamped oscillations. Note that
successive amplitudes are separated by time intervals of T /2. Suc-
cessive amplitudes of the same sign are separated by time intervals
of T .

We consider the time tn and tn+1 such that

tn+1 = tn + T /2 (1.54)

Then the respective amplitudes at tn and tn+1 are

An = A0 e−rtn
and An+1 = A0 e−rtn+1 = A0 e−r(tn +T /2) (1.55)

Therefore,

An A0 e−rtn
= −r(t +T /2)
= erT /2 = constant (1.56)
An+1 A0 e n

erT /2 = d is a constant of motion called decrement.

rT
loge d = =λ (1.57)
2
is called the logarithmic decrement.
Alternative definition of λ: Instead of considering successive ampli-
tudes (either positive or negative), only positive (or only negative) am-
plitudes may be considered. In this case, successive positive (or negative)

14
1.4 Parameters associated with underdamped harmonic oscillations

amplitudes are separated by a time difference of T , instead of T /2. Ac-


cording to this definition,

d = erT and λ = loge d = rT (1.58)

2. Relaxation time for amplitude (τA ): it is the time when the amplitude
of oscillations decreases to 1/e times its initial value.
1
If we put t = in the expression for A, then we get
r

A0
A = A0 e−r(1/r) = A0 e−1 = (1.59)
e

1
Thus, τA = is the time in which the amplitude reduces to the initial
r
amplitude, and hence is the relaxation time.
In terms of the relaxation time τA ,

A = A0 e−t/τA (1.60)

3. Relaxation time for energy (τ ): it is the time when the energy of


oscillations decreases to 1/e times its initial value.
From (1.52),

E = E0 e−2rt (1.61)

1
Therefore at t = ,
2r
E0
E = E0 e−2r1/2r = E0 e−1 = (1.62)
e

1
Thus, τ = is the relaxation time for energy of oscillations.
2r
In terms of the relaxation time τ ,

E = E0 e−t/τ (1.63)

4. Power dissipation (P ): it is the rate of decrease of energy due to

15
1.5 Forced oscillations

damping.
dE d 1 E
P =− = − (E0 e−t/τ ) = (E0 e−t/τ ) =
dt dt τ τ
E
Therefore, P = (1.64)
τ

5. Quality factor (Q): quality factor of an oscillator expresses its efficiency.


Energy stored per cycle
Q = 2π
Energy dissipated per cycle
E
= 2π where T is the time period of oscillations
PT
2π E
= ·
T P
= ωτ (1.65)

Thus, Q = ωτ .
For an ideal oscillator, τ → ∞ (no loss of energy). Therefore, in this
Q → ∞.

1.5 Forced oscillations


Objects can be forced to oscillate, most easily at their natural frequency.
Natural frequency is the frequency at which a system would oscillate if there
were no driving or damping forces (i.e., it is unaffected by external forces).
Applying a periodic driving force on a simple harmonic oscillator puts energy
into the system at the driving frequency.
When a body being acted upon by an external periodic force starts to oscillate
with the frequency of the external periodic force, the body is said to undergo
forced oscillations.

1.5.1 Equation of motion


Let an oscillator be subjected to an external periodic force, in addition to
restoring force and damping force. The forces acting on such an oscillator are:

1. Restoring force:

FR ∝ −y
or, FR = −ky (1.66)

16
1.5 Forced oscillations

2. Damping force: A retarding (decelerating) force that is proportional to


its velocity v.
dy
FD ∝ −
dt
dy
or, FD = −b (1.67)
dt

3. External periodic force:

Fext = F0 sin ωt (1.68)

where F0 and ω are the amplitude and (angular) frequency, respectively,


of the periodic driving force.

Hence, the equation of motion for a oscillator of mass m is

d2 y
m 2 = Fext + FR + FD
dt
d2 y dy
or, m 2 = F0 sin ωt − ky − b
dt dt
d2 y dy
=⇒ m 2 +b + ky = F0 sin ωt
dt dt
d2 y b dy k F0
=⇒ + + y= sin ωt (1.69)
dt2 m dt m m
b k F0
Let = 2r, = ω02 and = f0 . Then we get
m m m

d2 y dy
2
+ 2r + ω02 y = f0 sin ωt (1.70)
dt dt
This is the equation of motion for forced oscillations.
r is the damping constant, and ω0 is natural frequency of the oscillator.
f0 is the driving acceleration.

1.5.2 Solution of the equation and expression for the displacement


The complete solution of the equation for forced oscillations (1.70) consists of
two parts: a transient solution, and a steady-state solution.

17
1.5 Forced oscillations

1. Transient solution
It is the solution of the equation obtained by setting the RHS of (1.70)
to zero, i.e.,

d2 y dy
2
+ 2r + ω02 y = 0 (1.71)
dt dt
which is identical to (1.30) and represents damped harmonic motion. As
we have discussed in Section 1.2, the displacement represented by this
equation decays exponentially with time, and hence is not of interest when
considering forced oscillations over relatively long durations of time.

2. Steady-state solution Once the transient displacements given by (1.71)


have died down, we have the steady-state motion of the system under
forced oscillations. The steady-state solution is to be obtained by trial
method.
Physically, we would expect that the driven system will oscillate with the
frequency of the driving force, and there will be a phase difference between
the driving force and the driven system.
We thus consider a solution of the form

y = A sin(ωt − θ) (1.72)

which satisfies both the conditions. It represents oscillations with (angu-


lar) frequency ω, same as that of the driving force, with a phase difference
of θ. Here A is the amplitude of the forces oscillations and θ is the phase
difference between the driving force and the driven system.
From (1.72),
dy
= Aω cos(ωt − θ)
dt
d2 y
and 2
= −Aω 2 sin(ωt − θ) (1.73)
dt

Using (1.72) and (1.73) in (1.70), we get

−Aω 2 sin(ωt − θ) + 2rAω cos(ωt − θ) + ω02 A sin(ωt − θ)


= f0 sin ωt
= f0 sin[θ + (ωt − θ)]
= f0 sin θ cos(ωt − θ) + f0 cos θ sin(ωt − θ) (1.74)

Now, (1.74) has to be valid for all values of t. Hence, the coefficients of

18
1.5 Forced oscillations

sin(ωt − θ) and cos(ωt − θ) on the LHS and RHS must be separately equal
to each other. That is,

−Aω 2 + ω02 A = f0 cos θ (1.75)

and

2rAω = f0 sin θ (1.76)

Squaring and adding (1.75) and (1.76), we get

A2 (ω02 − ω 2 )2 + 4r2 ω 2 = f02


 
(1.77)

Therefore, from (1.77), the amplitude of driven oscillations in the steady-


state is given by
f0
A= p 2 (1.78)
(ω0 − ω 2 )2 + 4r2 ω 2

And hence, from (1.72), the steady-state solution of (1.70) is


f0
y=p 2 sin(ωt − θ) (1.79)
(ω0 − ω 2 )2 + 4r2 ω 2

The phase difference (θ) between the driving force and the driven system
may be obtained by taking the ratio of (1.75) and (1.76):
2rAω f0 sin θ
2 =
−Aω 2 + ω0 A f0 cos θ
 
−1 2rAω
=⇒ θ = tan (1.80)
−Aω 2 + ω02 A

When the driving frequency matches the natural frequency of the


driven system, that is when ω = ω0 , we have
 
2rAω π
θ = tan−1 = tan−1 (∞) = (1.81)
0 2

1.5.3 Conditions on the amplitude A


1. ω << ω0 (low frequency of driving force)

19
1.5 Forced oscillations

In this case, from (1.78),

f0 F0 /m F0
A≈ 2 = = (1.82)
ω0 k/m k
Thus, in this case, the amplitude of oscillations depends only on the restor-
ing force constant k.

2. ω >> ω0 (high frequency of driving force)


In this case, from (1.78),
f0
A≈ (considering small r << ω, i.e., low damping)
ω2
F0 /m
=
ω2
F0
= (1.83)
mω 2
Thus, the amplitude decreases with increasing mass and drive frequency.

3. ω = ω0 (amplitude resonance) The maximum amplitude of oscillations


occurs when the driving frequency matches the natural frequency of the
driven system. When the two exactly match, that is when ω = ω0 , from
(1.78) we have
f0
A= (1.84)
2rω0
Thus, in this case, amplitude is high for small r (low damping). For
r → 0, A → ∞.
Apparently this corresponds to the maximum possible amplitude Amax .
But in presence of damping, Amax does not occur exactly at ω = ω0 .
For Amax , the condition to be satisfied is
dA
=0 (1.85)

20
1.5 Forced oscillations

From (1.78), it is evident that a simpler condition for the same is


d  2
(ω0 − ω 2 )2 + 4r2 ω 2 = 0


=⇒ 2(ω02 − ω 2 )(−2ω) + 4r2 (2ω) = 0
=⇒ 4ω(−ω02 + ω 2 + 2r2 ) = 0
=⇒ either ω = 0 or (−ω02 + ω 2 + 2r2 ) = 0
(1.86)

ω = 0 is ignored because it implies there is no driving force. Thus, we


have,

(−ω02 + ω 2 + 2r2 ) = 0
=⇒ ω 2 = ω02 − 2r2
q
=⇒ ω = ω02 − 2r2 (1.87)

Therefore, the maximum amplitude Amax occurs at


q
ω = ω02 − 2r2 < ω0 (1.88)

in presence of non-zero damping.


For very low damping (very small r << ω0 ), Amax occurs at

ω ' ω0 (1.89)

This is the case of amplitude resonance.

1.5.4 Sharpness of resonance and Quality factor


1. Resonance When the frequency of the driving force approaches the nat-
ural frequency of the driven system, the amplitude of oscillations becomes
very large. A small-amplitude driving force can produce a large-amplitude
response. This phenomenon is known as resonance.
Examples of resonance are found in tuning forks, pushing a swing, soldiers
marching, NMR (nuclear magnetic resonance) etc.

2. Sharpness of resonance Amplitude at resonance is large for low damp-


ing and small for heavy damping. Further, for low damping, amplitude
drops rapidly as ω shifts away from ω0 . Thus the resonance is sharp. For
heavy damping, the decay of amplitude with ω is more gradual. Thus the

21
1.5 Forced oscillations

resonance is flat.
The variation of peak amplitude with damping is schematically repre-
sented in Figure 1.5. The figure also indicates the shift in the resonance
frequency farther away from ω0 with increasing damping, as expected
from (1.88).
Increasing damping also has the effect of shifting the phase difference θ
at resonance from 900 to lower values.

Figure 1.5: (a) Effect of damping on amplitude of forced oscillations (Source:


xmDemo). (b) FWHM and Q-factor (Source: researchgate).

22
1.6 Coupled oscillations

3. Quality factor (Q-factor) The spread ∆ω of the angular frequency at


half the maximum amplitude is known as full width at half maximum or
FWHM or bandwidth.
Quality factor or Q-factor is the defined as the ratio
ω0
Q= (1.90)
∆ω
Q-factor is a measure of the sharpness of resonance.
∆ω small =⇒ Q large =⇒ sharp resonance.
∆ω large =⇒ Q small =⇒ flat resonance.

4. Bandwidth and damping It can be shown that


1
∆ω = 2r = (1.91)
τ
1
where τ = is defined as the relaxation time of oscillations, analogous
2r
to the case of damped harmonic oscillations.

1.6 Coupled oscillations


1.6.1 Revisiting SHM (for small angles) - the simple pendulum
A simple pendulum consists of a spherical bob (considered a point mass) at-
tached to a string (considered massless). It is an idealized mathematical model
of a real pendulum. We consider a pendulum of mass m and length L, as shown
in Figure 1.6. The net force F acting on the pendulum is the resultant of the
following forces:

1. mg acting downwards.

2. Tension T in the string, acting along its length.

We consider the radial (in the direction of the length of string) and tangential
(in the direction of motion of bob) components of the net force F acting on the
pendulum.
The radial component of force is

Fr = mg cos θ − T (1.92)

The tangential component of force is

Fθ = −mg sin θ (1.93)

23
1.6 Coupled oscillations

Figure 1.6: A simple pendulum (Source: Physics StackExchange).

Now, the displacement x of the pendulum is always along the tangential direc-
tion, as shown in Figure 1.6. Thus, this tangential component of the force acts
along the direction of displacement and opposes it, as indicated in 1.6. So, this
represents the restoring force acting on the pendulum to cause the oscillatory
motion. Therefore,

d2 x
Fθ = m 2 (1.94)
dt
So, from (1.93) and (1.94),

d2 x
m 2 = −mg sin θ (1.95)
dt
(1.95) does not represent simple harmonic motion (SHM). Recall from (1.4)
that for SHM, the restoring force has to be proportional to the displacement,
but in this case, it is not.
However, if we consider small oscillations, then for small θ,

sin θ ' θ (1.96)

24
1.6 Coupled oscillations

Therefore, only for small θ,

Fθ = −mgθ (1.97)
x
Also, for small θ, θ ' tan θ ' . Therefore,
L
x
θ' (1.98)
L
Therefore, using (1.96) and (1.98) in (1.95), we get

d2 x x
m 2 = −mg (1.99)
dt L

Rearranging, we get

d2 x g
+ x=0 (1.100)
dt2 L
which is the differential equation representing simple harmonic motion of a
simple pendulum for small oscillations.

Comparing (1.100) with (1.4), we note that


r
2 g g
ω = or ω = (1.101)
L L
and hence, the corresponding time period of oscillations is
s
2π L
T = = 2π (1.102)
ω g

1.6.2 Coupled oscillations - coupled pendulums


Consider two identical pendulums of length l and supporting mass m, coupled
by a weightless spring of spring constant k and natural length equal to the
separation of the masses at zero displacement, as shown in Figure 1.7.

1. Equation of motion
Considering only small oscillations, equations of motion of the two masses,

25
1.6 Coupled oscillations

Figure 1.7: Two identical pendulums coupled via a massless spring (Source:
Matt Jarvis, Professor of Astrophysics and Fellow, St Cross Col-
lege).

with displacements x and y, respectively, are

d2 x x
m 2 = −mg − k(x − y) (1.103)
dt L

d2 y y
and, m 2 = −mg − k(y − x) (1.104)
dt L
or,

d2 x 2 k
+ ω 0 x = − (x − y) (1.105)
dt2 m
and
d2 y 2 k
+ ω 0 y = − (y − x) (1.106)
dt2 m

where the additional restoring force terms −k(x − y) and −k(y


r− x) are
g
due to coupling of the pendulums via the spring, and ω0 = is the
l
natural frequency of each pendulum.

26
1.6 Coupled oscillations

Adding (1.105) and (1.106), we get

d2
2
(x + y) + ω02 (x + y) = 0 (1.107)
dt

Subtracting (1.106) and (1.105), we get

d2 2 2k
(x − y) + ω 0 (x − y) = − (x − y)
dt2 m
d2
 
2 2k
or, (x − y) + ω 0 + (x − y) = 0 (1.108)
dt2 m

Replacing

X = x + y in (1.107)
and Y = x − y in (1.108) (1.109)

we get

d2 X
2
+ ω02 X = 0 (1.110)
dt
and
d2 Y
 
2k
2
+ ω02 + Y =0
dt m
d2 Y
or, 2
+ ω12 Y = 0 (1.111)
dt
(1.110) and
s(1.111) represent SHM with natural angular frequencies ω0

2k
and ω1 = ω02 + , respectively.
m

2. Solutions and resultant displacement


General solutions of (1.110) and (1.111), respectively, can be written as1

X = x + y = X0 cos(ω0 t + φ0 ) (1.112)

and

Y = x − y = Y0 cos(ω1 t + φ1 ) (1.113)

1
Note that we can choose to write the general solutions in terms of either sine or cosine functions. Here we
have opted for cosine functions, but the analysis holds equally good with sine functions.

27
1.6 Coupled oscillations

Based on the relation between x and y, different modes of coupled oscil-


lation may be observed, the most important of which are listed below.

1. In-phase oscillations
In this case x = y, or Y = x − y = 0. Both pendulums follow (1.110)
and oscillate with the angular frequency ω0 . The pendulums are always in
phase. The spring is always unstretched / uncompressed and maintains its
natural length. The relative motion of the pendulums and corresponding
time variation of x and y are schematically represented in Figure 1.8.

Figure 1.8: In-phase coupled oscillations (Sources: Matt Jarvis, Professor of As-
trophysics and Fellow, St Cross College and PPLATO@University
of Reading).

2. Out-of-phase oscillations
In this case x = −y, or X = x + y = 0. Both s pendulums follow (1.111)
 
2k
and oscillate with the angular frequency ω1 = ω02 + > ω0 . The
m
pendulums are always out of phase. The spring is always stretched /

28
1.6 Coupled oscillations

compressed. The relative motion of the pendulums and corresponding


time variation of x and y are schematically represented in Figure 1.9.

Figure 1.9: Out-of-phase coupled oscillations (Sources: Matt Jarvis, Pro-


fessor of Astrophysics and Fellow, St Cross College and
PPLATO@University of Reading).

These two modes are called normal modes of oscillation. ω0 and ω1 are
called normal frequencies. Any other frequency is a linear combination
of ω0 and ω1 .

3. Resonance
In general, from (1.112) and (1.113),
1
x = (X + Y )
2
1
and, y = (X − Y ) (1.114)
2

29
1.6 Coupled oscillations

Consider the special condition where the resultant amplitudes and phases
given by (1.112) and (1.113) are equal. That is,

X0 = Y0 = a (say)
and φ0 = φ1 = 0 (the equal value can be zero without loss of generality)
(1.115)

Then, the displacement of the first pendulum


1
x = (X + Y )
2
= a cos ω0 t + a cos ω1 t
(ω1 − ω0 )t (ω1 + ω0 )t
= 2a cos cos (1.116)
2 2
and the displacement of the second pendulum
1
y = (X − Y )
2
= a cos ω0 t − a cos ω1 t
(ω1 − ω0 )t (ω1 + ω0 )t
= 2a sin sin (1.117)
2 2
Clearly, the resultant displacements x and y are a superposition of two
different frequencies; a higher frequency
(ω1 + ω0 )
ωC = (1.118)
2
with a corresponding shorter time period

TC = (1.119)
ω 1 + ω0
and a lower frequency
(ω1 − ω0 )
ωB = (1.120)
2
with a corresponding longer time period

TB = (1.121)
ω1 − ω0

ωB is called Beats frequency.

30
1.6 Coupled oscillations

Figure 1.10: Resonance in coupled oscillations (Source: PPLATO@University


of Reading).

We can also define the In-phase time period



T0 = (1.122)
ω0
and the Out-of-phase time period

T1 = < T0 (1.123)
ω1

Then from (1.119), (1.121), (1.122), and (1.123), we have


T0 T1
TC = 2 (1.124)
T0 + T1
and
T0 T1
TB = 2 (1.125)
T0 − T1

Degree of coupling is defined as

ω12 − ω02 T02 − T12


χ= 2 = 2 (1.126)
ω1 + ω02 T0 + T12

With the initial conditions x = 2a, y = 0 at t = 0, the displacements of

31
1.6 Coupled oscillations

the coupled pendulums are as schematically represented in Figure 1.10.


In case of resonance, there is continuous transfer of energy between the
two coupled oscillators, as indicated by the complementary changes in
their respective displacements in Figure 1.10.

Figure 1.11: Spring-mass coupled oscillator system (Source: OSU).

Exercise: Work out the details of coupled oscillations for a spring-mass


system, as shown in Figure 1.11, instead of coupled pendulums discussed above.
That is, set up the equations of motion of the spring-coupled masses, and
discuss the normal modes of oscillations. And thus verify that similar coupled
oscillation features are observed in both systems.

1.6.3 Oscillations in practice


The simple harmonic oscillator is only a mathematical construct to enable un-
derstanding of the physics of oscillations. Practically all oscillators in real-life
are damped or forced.
The most common example of an underdamped oscillator is the simple pen-
dulum. Another good example is a mass-spring system. The swing (as seen in
parks) involves forced oscillations (by someone pushing it). Once the external
force is removed, its motion is damped (just like the pendulum).
Overdamping is typically used in door dampers (and similar dampers) as the
system goes to equilibrium (without overshooting it) over a sufficiently long
time. A shock absorber is essentially a damped spring oscillator, the damping
is from a piston moving in a cylinder filled with oil. If the oil is really thick, or
the piston too tight, the shock absorber will be too stiff - it won’t absorb the
shock, and you will! This is the case of overdamping.
We need to tune the damping so that the car responds smoothly to a bump
in the road, but does not continue to bounce after the bump. this is achieved
using critical damping.
Also note that although we have confined our discussion to mechanical os-
cillations only, oscillations are not only of mechanical nature, but can also be

32
1.7 Waves

electromagnetic. Mechanical oscillations are characterized by alternating con-


versions of the kinetic energy into one (or several) kinds of potential energy and
back. In electromagnetic oscillations, alternating conversions occur between the
electric field energy (which is analogous to the potential energy in mechanical
systems) and the magnetic field energy (the analogue of the kinetic energy).
Sometimes oscillations have a combined mechanical and electromagnetic na-
ture, e.g., oscillations in plasma.

1.7 Waves
1.7.1 Introduction
Wave motion is periodic motion in space and time. A disturbance of any
conventional property y of a medium may spread through space with time.
Such a spreading of disturbances in a medium is called a wave. y may be
called the wave function.
Let y represent displacement of particles in the medium. Then wave motion
can result from periodic displacement of particles from their mean positions.
The state of motion of a particle is called phase. Different particles may be
in different phases at a given time.

Wave motion consists of the propagation of phase from point to point in the
medium, distinct from the motion of particles in the medium2 .

Surfaces of constant phase in a medium are called wavefronts. A wave


whose wavefronts are infinite parallel planes are called plane waves.

Types of waves:

1. Transverse waves are waves in which the displacement of particles about


their mean position is perpendicular to the direction of propagation. Ex-
amples of transverse waves are EM (electromagnetic) waves and waves in
a pulled string.

2. Longitudinal waves are those in which the displacement of particles


about their mean position is along the direction of propagation. Examples
of longitudinal waves are sound waves and propagation of compressions
and rarefactions in a slinky.

33
1.7 Waves

Figure 1.12: (a) Longitudinal and (b) transverse waves demonstrated using a
slinky.

In the next Sections, we will focus on transverse waves; discuss the important
parameters, and set up the general wave equation and its differential form.

1.7.2 Parameters associated with a transverse wave

Figure 1.13: Amplitude, wavelength, and time period of a simple harmonic


transverse sinusoidal wave (Source: sengpielaudio).

1. Amplitude (a): maximum displacement of a particle from its mean


position.
2. Time period (T ): time taken to complete one wave cycle, as shown in
Figure 1.13.

34
1.7 Waves

1
3. Frequency (n or ν): number of wave cycles per second. n = .
T

4. Angular frequency (ω): ω = 2πn = .
T
5. Wavelength (λ): distance travelled by the wave in one cycle, i.e., in
time T , as shown in Figure 1.13.

6. wave number (k): a measure of the number of wave cycles per unit

distance. k = .
λ
λ 1
7. Velocity (v): v = . Since n = , we have
T T
v =n·λ (1.127)

1.7.3 Wave equation (transverse wave)


A simple harmonic progressive wave advances continuously in a given direction
without change in form and particles of the medium perform simple harmonic
oscillations about their mean position with the same amplitude and time period.

Figure 1.14: A simple harmonic progressive wave (Source: AnkPlanet).

We consider a simple harmonic progressive transverse wave as schematically


represented in Figure 1.14. At any time t, the displacements of a particle at
the origin O is given by

y = a sin ωt (1.128)

35
1.7 Waves

Consider another particle at P , at a distance x from O. Then the displacements


at P are given by

y = a sin(ωt − φ) (1.129)

where φ is the phase difference between O and P .


Now, for one complete wave cycle, i.e., for a particle at B, at a distance λ
from O as shown in Figure 1.14,

x=λ
and φ = 2π (1.130)

because over a complete wave cycle, the phase goes through a cycle of 0 to 2π.
Therefore, from (1.130),
2πx
φ= = kx (1.131)
λ
If v is the wave velocity, then
2πv
ω = 2πn = (1.132)
λ
Using these expressions for φ and ω in (1.129), we can write
 
2πvt 2πx
y = a sin − (1.133)
λ λ
or,


y = a sin (vt − x) (1.134)
λ

For a particle at a distance x in the negative direction from O, then displace-


ment is given by


y = a sin (vt + x) (1.135)
λ

We can rewrite (1.134) and (1.135) in terms of ω and k as

36
1.7 Waves

y = a sin(ωt − kx) (for positive x) (1.136)

and

y = a sin(ωt + kx) (for negative x) (1.137)

1.7.4 Differential equation of wave motion


The displacement for a simple harmonic progressive transverse wave is given by
(1.136)

y = a sin(ωt − kx) (1.138)

Differentiating3 (1.138) w.r.t. t, twice, we get


∂y
= aω cos(ωt − kx)
∂t
∂ 2y
=⇒ 2
= −aω 2 sin(ωt − kx) (1.139)
∂t
Differentiating (1.138) w.r.t. x, twice, we get
∂y
= −ak cos(ωt − kx)
∂x
∂ 2y
=⇒ 2
= −ak 2 sin(ωt − kx) (1.140)
∂x
Comparing (1.139) and (1.140), we have

∂ 2y ω2 ∂ 2y
= 2· 2 (1.141)
∂t2 k ∂x
Now,
ω 2πn
= =n·λ=v (1.142)
k 2π/λ

Therefore, from (1.141) and (1.142),

3
Note that here we switch to partial derivatives from total derivatives. This is because, unlike in case of
oscillations, where the displacement is a function of time (t) only, in case of waves the displacement is a
function of both time (t) and position (x).

37
1.7 Waves

∂ 2y 2
2 ∂ y
=v · 2 (1.143)
∂t2 ∂x
which is the differential equation of motion for a transverse wave travelling with
velocity v.

38
1.7 Waves

Note: Initial phase at the origin and other equivalent forms

Note that, in the above treatment, we have considered that the phase of the
particle at the origin is zero. This does not affect the general applicability of
the wave equation (1.134) or (1.136). However, one may consider some non-
zero initial phase φ0 at the origin, and in this case, the further generalized wave
equation is

y = a sin(ωt − kx − φ0 ) (1.144)

Further note that, instead of (1.134) and (1.136), we can also write the wave
equation as

y = a sin (x − vt) = a sin(kx − ωt) (1.145)
λ
.

1.7.5 Numerical problems

Example 1: The equation of motion of a wave travelling in a spring is


y = 4 sin π(0.10x − 2.0t). Find the (i) amplitude, (ii) wavelength, (iii) initial
phase at the origin, (iv) speed, and (v) frequency of the wave. Here time is
given in seconds and distance in cm.

Solution:

To determine the required parameters, we must rewrite the given equation


in a form similar to the general equation and compare term-by-term.
The general wave equation similar in form to the given equation is

y = a sin (x − vt)
λ
We rearrange the given equation as

y = 4 sin[(0.1π)(x − 20t)]

Comparing term-by-term:
(i) amplitude, a = 4 cm.

(ii) = 0.1π =⇒ wavelength λ = 20 cm.
λ
(iii) Initial phase φ0 = 0.

39
1.7 Waves

(iv) v = 20 cm/s.
v 20
(v) Frequency n = = = 1.0 Hz.
λ 20

Example 2: A radio station broadcasts at a frequency of 15 × 106 Hz.


Calculate the length of the radio waves, given that velocity of radio waves is
equal to the velocity of light, i.e., 3 × 108 m/s.

Solution:

Given, frequency of the radio waves, n = 15×106 Hz, and velocity v = 3×108
m/s.
Therefore, length of the waves (i.e., the wavelength)
v 3 × 108
λ= = = 20 m.
n 15 × 106
********************* The End (of this Unit) *********************

40
Interference of light
PH-1007 (Physics)
Dr. Gorky Shaw

2.1 Introduction
Light is a transverse, electromagnetic wave that is visible to the human eye.
Light is dualistic in nature.
Certain optical phenomena, such as interference, diffraction, and polarization,
are explained by the wave nature of light. Light-matter interaction in processes
of emission and absorption, such as photoelectric effect and Compton effect,
are explained by the corpuscular (particle) nature of light.

2.1.1 A few terms


1. Huygens’ principle
Every point on a wavefront is a source of wavelets, which spread forward
at the same speed.

Figure 2.1: Schematic representation of Huygens’ principle (a) Secondary


wavelets in a plane wavefront (b) Secondary wavelets emanating
from a small opening, acting as a point source (Sources: OSU and
HyperPhysics ).

2. Superposition principle
When two waves interact, the resulting wave function is the sum of the
two individual wave functions.
2.1 Introduction
Consider two waves with wave functions y1 and y2 , represented by the
equations

y1 = a1 sin(ω1 t − k1 x)
and y2 = a2 sin(ω2 t − k2 x) (2.1)

When these two waves interact, the resultant wave function y is given by

y = y1 + y2
= a1 sin(ω1 t − k1 x) + a2 sin(ω2 t − k2 x) (2.2)

In general, for n number of interacting waves, the resultant wave function


is given by

y = y1 + y2 + y3 + ... + yn
Xn
= yi (2.3)
i=1

3. Coherent sources
Two light sources are said to be coherent if they produce waves that have
the same frequency, and a sharply defined phase difference that remains
unchanged with time. The most common example of coherent sources is
Laser.

4. Interference
When two waves produced by coherent sources travel simultaneously in
a medium and superpose each other, the resultant intensity is not dis-
tributed uniformly in space. This modification in intensity is called in-
terference.

5. Constructive and destructive interference


At points where a crest (trough) of one wave falls on a crest (trough) of
the other, the resultant amplitude is the sum of the amplitudes of each
wave separately. The intensity at these points is maximum. This is the
case of constructive interference.
At points where a crest of one wave falls on a trough of the other, the
resultant amplitude is the difference of the amplitudes of each wave. The
intensity at these points is minimum. This is the case of destructive
interference.

2
2.1 Introduction

Figure 2.2: (a) Constructive and (b) destructive interference, a direct conse-
quence of the superposition principle (Source: A-Level Physics).

2.1.2 Conditions for the interference of light


For sustained interference:
1. The sources must be coherent, i.e., they must maintain a constant phase
difference between them.
2. The two waves must have the same frequency (and hence the same wave-
length). This also means the sources should be monochromatic, i.e., of a
single wavelength.
3. Polarized waves must be in the same state of polarization.
For observation:
1. The separation between the sources should be small.

3
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit
2. The distance of observation should be large.
For good contrast:
1. Amplitudes of the interfering waves should be nearly equal (and ideally,
exactly equal).

2. The sources must be narrow.

2.1.3 Types of interference


1. Division of wavefront by employing mirrors, biprisms, or lenses. Divi-
sion of wavefront requires a point source or narrow slit source.
Examples: Fresnel biprism, Young’s double slits, Fresnel’s double mirror,
Lloyd’s mirror.

2. Division of amplitude of incident beam into two or more parts, either


by partial reflection or refraction, which traverse different paths and are
finally brought together to produce interference.
Examples: Interference in thin films, Newton’s rings, Michelson’s inter-
ferometer.

2.2 Analytical treatment of interference by division of


wavefront - Young’s double slit
The wave nature of light was strongly inferred by an experiment performed by
Thomas Young. It was based on how light waves passing through two closely
spaced slits interact. Figure 2.3 shows a schematic representing the processes
in his experiment illustrating interference by division of wavefront. Next we
discuss the observations analytically.
As shown in Figure 2.4, let S be narrow slit illuminated by a monochromatic
light source, and S1 and S2 two similar parallel slits close to each other and
equidistant from S. Then S1 and S2 become centres of secondary wavelets and
act as coherent sources.

2.2.1 Resultant amplitude and intensity


Let a1 and a2 be the amplitudes at P due to the waves from S1 and S2 , respec-
tively. The waves reaching P will have different path lengths S1 P and S2 P.
Hence, path difference between the two waves reaching P from S1 and S2

= S2 P − S1 P (2.4)

4
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit

Figure 2.3: Young’s double slit experiment (Source: University of Texas).

Corresponding phase difference at P,



δ= × (S2 P − S1 P) (2.5)
λ
where λ is the wavelength of light used.
We may write the individual displacements at P, with a phase difference of δ
between them as

y1 = a1 sin ωt
and y2 = a2 sin(ωt + δ) (2.6)

where ω is the common angular frequency of the waves.


By the principle of superposition, the resultant displacement at P,

y = y1 + y2
= a1 sin ωt + a2 sin(ωt + δ)
= a1 sin ωt + a2 (sin ωt cos δ + cos ωt sin δ)
= (a1 + a2 cos δ) sin ωt + (a2 sin δ) cos ωt (2.7)

5
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit

Figure 2.4: Schematic for analytical treatment of interference by division of


wavefront.

We make the replacements with new variables R and θ given by

a1 + a2 cos δ = R cos θ
and a2 sin δ = R sin θ (2.8)

Then, from (2.7) and (2.8),

y = R cos θ sin ωt + R sin θ cos ωt


or, y = R sin(ωt + θ) (2.9)

(2.9) indicates two notable features of the resultant wave: its frequency re-
mains the same (angular frequency ω same as that of the source waves), and its
nature remains unaltered (similar forms of the wave equations (2.6) and (2.9)).
Here, R is the resultant amplitude. Hence the resultant intensity is I = R2 .
Squaring and adding the two equations in (2.8), we get

I = R2 = a21 + a22 + 2a1 a2 cos δ (2.10)

6
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit

2.2.2 Conditions for maxima and minima


From (2.10), intensity at P is maximum if

cos δ = +1
=⇒ δ = 2nπ, n = 0, ±1, ±2, ... (2.11)

And corresponding path difference


λ
= S2 P − S1 P = × δ = nλ (2.12)

Maximum value of intensity,

Imax = a21 + a22 + 2a1 a2


= (a1 + a2 )2
> a21 + a22 = I1 + I2 (2.13)

That is,

Imax > I1 + I2 (2.14)

where I1 and I2 are the intensities of the source waves at P.

Special case: equal source amplitudes

If a1 = a2 = a and I = a2 , then

Imax = 4a2 = 4I (2.15)

On the other hand, intensity at P is minimum if

cos δ = −1
=⇒ δ = (2n + 1)π, n = 0, ±1, ±2, ... (2.16)

And corresponding path difference


λ λ
= S2 P − S1 P = × δ = (2n + 1) (2.17)
2π 2

7
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit

Minimum value of intensity,

Imin = a21 + a22 − 2a1 a2


= (a1 − a2 )2
< a21 + a22 = I1 + I2 (2.18)

That is,

Imin < I1 + I2 (2.19)

where I1 and I2 are the intensities of the source waves at P.

Special case: equal source amplitudes

If a1 = a2 = a and I = a2 , then

Imin = 0 (2.20)

2.2.3 Interference and conservation of energy

Figure 2.5: Variation of resultant intensity (I) with phase difference (δ).

To determine the average intensity of the resultant wave, we need to deter-


mine the average intensity over a single complete cycle, i.e., between δ = 0 and

8
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit

δ = 2π, which is given by


R 2π
Idδ
Iavg = R0 2π
0 dδ
R 2π 2 2
0 (a1 + a2 + 2a1 a2 cos δ)dδ
= R 2π
0 dδ
2π(a21 + a22 )
=

= a1 + a22
2

= I1 + I2 (2.21)

Hence, although there is a variation of resultant intensity between Imax and


Imin , the average resultant intensity Iavg is equal to the sum of the intensities
of the source waves. This means, there is no loss or gain of energy due to
interference. There is just a redistribution of the energy from the source waves.

2.2.4 The (not-so-obvious) small angle approximation


In the analytical treatment above, there is no explicit mention of (i) the small
separation between the sources, or (ii) large distance of observation, both of
which are mentioned earlier as necessary conditions for observation of interfer-
ence pattern. These are indeed incorporated, but in a subtle manner.
Expression (2.8) for the resultant displacement at P, based on the superpo-
sition principle, is valid only if the paths S1 P and S2 P traversed by the source
waves are parallel to each other. As evident from Figure 2.4, in general, the two
paths would not be parallel. Only if the separation between the sources is much
smaller than the distance of observation, i.e., if D >> 2d, can we consider the
two paths S1 P and S2 P to be approximately parallel.

9
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit

2.2.5 Fringe width in double slit interference


The bright and dark bands produced as a result of interference of light are
known as interference fringes. We may estimate the width of these fringes
as discussed below.
In Figure 2.4, we have

(S2 P)2 = (S2 M2 )2 + (PM2 )2


= D2 + (x + d)2
(x + d)2
 
2
=D 1+
D2
"  2 #1/2
x+d
=⇒ S2 P = D 1 + (2.22)
D

We consider only small distances x << D. And as discussed above, for


observation of interference pattern, we must have d << D. Hence, we have
(x + d) << D. So, in the above expression, we can use binomial expansion and
keep the first two terms of the expansion, ignoring higher order terms1 , we get
"  2 #
1 x+d
S2 P = D 1 +
2 D
(x + d)2
=D+ (2.23)
2D
Similarly,

(x − d)2
S1 P = D + (2.24)
2D
Therefore,
2xd
S2 P − S1 P = (2.25)
D
For maxima or bright fringes:

   
1 n
P∞ n k n n(n − 1)...(n − k + 1)
For a real number n, the binomial expansion of (1+x) = k=0 k x , where k = .
k!
Here k is a positive integer.

10
2.2 Analytical treatment of interference by division of wavefront - Young’s
double slit

2xd
= nλ
D

=⇒ x = n · (2.26)
2d
For minima or dark fringes:

2xd λ
= (2n + 1)
D 2
(2n + 1) Dλ
=⇒ x = · (2.27)
2 2d
Therefore, spacing between two consecutive bright (or dark) fringes

= xn+1 − xn

= (2.28)
2d
which is independent of n.

It is called fringe width (β).



β= (2.29)
2d

One may define the angular fringe width βθ as


β λ
βθ = = (2.30)
D 2d

11
2.3 Numerical problems - Interference by division of wavefront

2.3 Numerical problems - Interference by division of


wavefront

Exercise 2.1
Two coherent sources with intensity ratio 100:1 produce interference fringes.
Find the ratio between maximum and minimum intensities in the interference
pattern.

Solution:

We know,

Imax (a1 + a2 )2
=
Imin (a1 − a2 )2
Now, given,

I1 a21 100
= 2=
I2 a2 1
a1
=⇒ = 10
a2
=⇒ a1 = 10a2

Therefore,

Imax (11a2 )2 121


= =
Imin (9a2 )2 81

Exercise 2.2
The ratio between maximum and minimum intensities in a double slit inter-
ference pattern is 36:1. Find the ratio between amplitudes and intensities of
the two interfering waves.

Solution:

12
2.3 Numerical problems - Interference by division of wavefront

Given,

Imax (a1 + a2 )2 36
= =
Imin (a1 − a2 )2 1
=⇒ a1 + a2 = 6(a1 − a2 )
=⇒ (a1 + a2 ) = 6(a1 − a2 )
=⇒ 5a1 = 7a2

Therefore, ratio of amplitudes of the interfering waves,


a1 7
=
a2 5
and the ratio of their intensities,

I1 a21 49
= 2=
I2 a2 25

Exercise 2.3
In a Young’s double slit experiment with monochromatic light of wavelength
6000 Å, the fringe width is found to be 0.5 mm in the interference pattern on a
screen at a distance 1 m from the slits. Find the separation between the slits.

Solution:

Given, β = 0.5 mm = 0.05 cm,


λ = 6000 Å= 6 × 10−7 cm,
D = 1 m = 100 cm.
Therefore, separation between the slits,

Dλ 100 × 6 × 10−7
2d = = = 0.0012 cm
β 0.05

Exercise 2.4
Two coherent sources placed 0.2 mm apart produce an interference pattern
observed on a screen 1 m away. With a certain monochromatic light source, the
fourth bright fringe is situated at a distance 10.0 mm from the central fringe.
Find the wavelength of light used.

13
2.3 Numerical problems - Interference by division of wavefront

Solution:

Position of the nth bright fringe,



xn = n ·
2d
Here,
n = 4,
D = 1 m = 1000 mm,
2d = 0.2 mm,
xn = 10 mm.
Therefore,
2dxn
λ=
nD
0.2 × 10
=
4 × 1000
= 5 × 10−4 mm
= 5000 Å

Exercise 2.5
In an interference pattern, 12th order maximum is observed for λ = 6000 Å.
What order maximum is visible with light of wavelength 4800 Å?

Solution:

We know, the path difference between source waves for the nth order maxi-
mum ∝ nλ.
Given, maxima (of different orders) for two wavelengths occur at the same
position, or

n2 λ2 = n1 λ1
n1 λ1 12 × 6000
or, n2 = = = 15
λ2 4800

14
2.4 Interference due to thin films

2.4 Interference due to thin films


2.4.1 Phase change on reflection - Stokes’ treatment
When a light wave is reflected at the surface of an optically denser medium
(with higher refractive index), it suffers a phase change of π (but not when
reflected at the surface of a rarer medium (with lower refractive index)).
This is illustrated in Figure 2.6.

Figure 2.6: Phase change on reflection (Source: Stack Exchange).

2.5 Interference due to parallel-sided thin films: reflected


rays
Let a monochromatic light beam SA be incident at angle i on a parallel-sided
transparent thin film of thickness t and refractive index µ > 1, as shown in
Figure 2.7. At A, it is partly reflected along AR1 at angle i, and partly refracted
along AB at angle r. At B, it is again partly reflected along BC and partly
refracted along BT1 . Similar reflections and refractions occur at C, D etc.
Path difference between the rays AR1 and CR2 ,

p = µ(AB + BC) − AN (2.31)

15
2.5 Interference due to parallel-sided thin films: reflected rays

Figure 2.7: Interference by division of amplitude due to a parallel-sided thin


film.

But,

AB = BC = BM sec r = t sec r (2.32)

and

AN = AC sin i = (AM + MC) sin i


= 2AM sin i (since AM = MC)
= 2BM tan r sin i (since AM = BM tan r)
= 2t tan r sin i
sin i
= 2t tan r(µ sin r) (since = µ)
sin r
= 2µt sec r sin2 r (2.33)

16
2.5 Interference due to parallel-sided thin films: reflected rays

Therefore,

p = 2µt sec r − 2µt sec r sin2 r


= 2µt sec r cos2 r
= 2µt cos r (2.34)

Further, the ray AR1 undergoes an additional phase difference of π, and


λ
hence, and additional path difference of , where λ is the wavelength of light
2
used.
Therefore, effective path difference between AR1 and CR2
λ λ
peff = p ± = 2µt cos r ± (2.35)
2 2
Thus, depending on the values of t and λ, the effective path difference between
the reflected rays can lead to constructive or destructive interference, and the
film may appear dark or bright in reflected light.
We can use either of the + and − signs. In the following discussion, we use
the + sign.

2.5.1 Conditions for maxima and minima

For maximum intensity of reflected light,


λ
2µt cos r + = nλ, n = 1, 2, 3...
2
λ
or, 2µt cos r = (2n − 1) (2.36)
2
In this condition, the thin film will appear bright in reflected light.
For minimum intensity of reflected light,
λ λ
2µt cos r + = (2n + 1) , n = 0, 1, 2, 3...
2 2
or, 2µt cos r = nλ (2.37)

In this condition, the thin film will appear dark in reflected light.

Note that the choice of + and − signs, as well as values of n (0, 1, 2... or 1, 2, 3...)
in the conditions (2.36) and (2.37) are not arbitrary. These are carefully chosen

17
2.6 Interference due to thin films - further points

so that the value of t, the thickness of the film, is always positive. Negative
value of t would not make physical sense.

2.6 Interference due to thin films - further points


2.6.1 Interference due transmitted rays
There is no additional phase change due to reflection at points like B and C in
Figure 2.7 because, in these cases, light is travelling from the denser medium
to the rarer medium.
For maximum intensity of transmitted light,

2µt cos r = nλ (2.38)

For minimum intensity of reflected light,


λ
2µt cos r = (2n − 1) (2.39)
2

2.6.2 Colours in thin films


White light is composed of light of different wavelengths. Since every colour
of light has a different wavelength, according to (2.36), only one colour can
constructively interfere at a given thickness of a film. Hence, different colours
appear brighter for different thicknesses of films.
Some common examples of this phenomenon are the rainbow colour patterns
seen in soap bubbles, oil spills, bottoms of metallic cooking pots etc. (see Figure
2.8).

2.6.3 Interference due to wedge-shaped thin films

Figure 2.9: Wedge-shaped thin film.

18
2.6 Interference due to thin films - further points

Figure 2.8: Colours in thin films (a) Soap bubbles, (b) Oil spill, and (c) Bottom
of a cooking pot.

For a wedge-shaped thin film with wedge angle θ, as shown in Figure 2.9, the
effective path difference at thickness t is given by
λ
peff = 2µt cos (r + θ) ± (2.40)
2
Accordingly, in reflected light, for maximum intensity,
λ
2µt cos (r + θ) = (2n − 1) , n = 1, 2, 3... (2.41)
2
and for minimum intensity,

2µt cos (r + θ) = nλ, n = 0, 1, 2, 3... (2.42)

For such a film, the thickness varies across the film. Hence, alternate locations
will demonstrate constructive and destructive interference, according to the
thickness of the film at that location. This is the basis of Newton’s rings
formation, which is discussed next.

19
2.7 Newton’s rings

2.7 Newton’s rings


Newton’s rings is a phenomenon in which interference pattern is created by the
reflection of light from an air film formed between a spherical surface and an
adjacent touching flat surface.
With monochromatic light, a series of concentric alternate bright and dark
rings are observed. With white light, concentric ring patterns of rainbow colours
is observed, because the different wavelengths of light interfere constructively
at different thicknesses of the air film.

Figure 2.10: Experimental setup for observation of Newton’s rings.

A typical experimental setup for observation of Newton’s rings is schemati-


cally represented in Figure 2.10. Usually, monochromatic light is made incident
normally on a combination of a plano-convex lens and a plane glass plate, as
shown in the Figure. The reflected light is observed using a travelling micro-
scope. A thin air film of variable thickness is formed between the lens and the
glass plate.
Alternate bright and dark circular rings, with a central dark spot, are ob-
served in the reflected light. This Newton’s rings formation is a result of in-

20
2.7 Newton’s rings

terference between light waves reflected from the upper and lower surfaces of
the wedge-shaped air film. The thickness of the air film is zero at the point of
contact and gradually increases outwards. Due to the shape of the plano-convex
lens, the locus of points where the thickness of the air film is constant is a circle,
with the point of contact as its centre.

2.7.1 Effective path difference and the central dark spot


The air film can be considered a wedge-shaped thin film. Then the effective
path difference between waves reflected from the upper and lowers surfaces at
thickness t is given by
λ
peff = 2µt cos (r + θ) + (2.43)
2
Now, for large radius of curvature of the plano-convex lens, θ can be ignored.
Further, for normal incidence, r = 0. Therefore, we have
λ
peff = 2µt + (2.44)
2
Therefore, similar to the conditions in Section 2.5.1, for maximum intensity
(bright rings),
λ
2µt = (2n − 1) , n = 1, 2, 3... (2.45)
2
and for minimum intensity (dark rings),

2µt = nλ, n = 0, 1, 2, 3... (2.46)

At the point of contact, t = 0. Hence, from (2.44), at this location,


λ
peff = (2.47)
2
for all λ. This is the condition for minimum intensity. Hence, the central spot
is always dark.

2.7.2 Diameter of rings


Consider the plano-convex lens LOL’ placed on a glass plate AB, the point of
contact being O, as shown in Figure 2.11. Let R be the radius of curvature of

21
2.7 Newton’s rings

Figure 2.11: Determination of diameter of Newton’s rings.

the lens. Let rn be the radius, and Dn = 2rn the diameter, of the Newton’s ring
corresponding to a point P where the film thickness is t.
We use the property of a circle2 :

rn2 = PN2 = ON × NE
= t × (2R − t)
= 2Rt − t2 (2.48)

Considering R >> t, we have

rn2 = 2Rt
r2
=⇒ 2t = n (2.49)
R

2
Verify this yourself!

22
2.7 Newton’s rings

Therefore, from (2.45) and (2.49), for a bright ring (constructive interference),

rn2 λ
µ = (2n − 1) , n = 1, 2, 3...
R 2
λR
=⇒ rn2 = (2n − 1) (2.50)

That is, for bright rings,


λR
Dn2 = 2(2n − 1)
µ
p
or, Dn ∝ (2n − 1) (2.51)

From (2.46) and (2.49), for a dark ring (destructive interference),

rn2
µ = nλ, n = 0, 1, 2, 3...
R
nλR
=⇒ rn2 = (2.52)
µ

That is, for dark rings,


4nλR
Dn2 =
µ

or, Dn ∝ n (2.53)

For air film, µ = 1. Therefore,


2
Dn[bright] = 2(2n − 1)λR
2
and Dn[dark] = 4nλR (2.54)

2.7.3 Determination of wavelength of monochromatic light using


Newton’s rings
From (2.54), for an air film, the diameter of the mth dark ring is given by
2
Dm = 4mλR (2.55)

23
2.7 Newton’s rings

and that of the (m + p)th ring is


2
Dm+p = 4(m + p)λR (2.56)

Therefore,
2 2
Dm+p − Dm = 4pλR (2.57)

That is,

2 2
Dm+p − Dm
λ= (2.58)
4pR

In the experimental setup shown in Figure 2.10, the cross-wire of the eye-
piece of the microscope is focused on any dark ring (say 24th, 16th etc.), and
reading of the microscope is noted. Similar readings are noted by focusing the
cross-wire after every few dark rings. Diameters of different rings are calculated,
and a graph is plotted between the square of the diameter (Dn2 ) and the ring
number n, which is a straight line, as shown in Figure . Its slope gives the value
2 2
Dm+p − Dm
of . If R is known, the wavelength λ of the monochromatic light
p
can then be determined from 2.58.

Figure 2.12: (Dn2 ) versus n plot.

24
2.8 Numerical problems - Interference by division of amplitude

2.7.4 Determination of refractive index of a liquid using Newton’s rings


For air film, from (2.57),
2 2
[Dm+p − Dm ]air = 4pλR (2.59)

If a liquid is put in between the plano-convex lens and the glass plate, then for
the liquid film,

2 2 4pλR
[Dm+p − Dm ]liquid = (2.60)
µ
Therefore, from (2.59) and (2.60),
2 2
[Dm+p − Dm ]air
µ= 2 2]
(2.61)
[Dm+p − Dm liquid

Hence, refractive index (µ) of the liquid can be determined by measuring the
diameters of Newton’s rings in the liquid and in air.

2.8 Numerical problems - Interference by division of


amplitude

Exercise 2.6
A soap film (µ = 1.33) seen by sodium light (λ = 5893 Å) by normal reflection
appears dark. Find the minimum thickness of the film.

Solution:

For the film to appear dark, i.e., for destructive interference, we have

2µt cos r = nλ

Clearly, the minimum possible value of the thickness of the film, t, is obtained
for n = 1.
For normal incidence, r = 0, cos r = 1
Therefore,
nλ 1 × 5893
t= = = 2215 Å
2µ 2 × 1.33

25
2.8 Numerical problems - Interference by division of amplitude

Exercise 2.7
Newton’s rings are observed in reflected light of wavelength 6250 Å. Diameter
of the 10th dark ring is 0.5 cm. Find (i) the radius of curvature of the lens (ii)
thickness of the air film.

Solution:

For air film, diameter of the nth dark ring, Dn , is given by

Dn2 = 4nλR

Given, n = 10, Dn = 0.5 cm, λ = 6250 Å= 6.25 × 10−5 cm.


(i) Therefore, radius of curvature of the lens

Dn2 (0.5)2
R= =
4nλ 4 × 10 × 6.25 × 10−5
0.25
=
250 × 10−5
= 100 cm
= 1m

(ii) Condition for dark ring:

2t = nλ

=⇒ t =
2
10 × 6.25 × 10−5
=
2
= 31.25 × 10−5
= 3.125 µm

Exercise 2.8
In a Newton’s rings experiment, the diameter of the 15th ring is found to
be 0.590 cm and that of the 5th ring 0.336 cm. If radius of curvature of the
plano-convex lens is 100 cm, calculate the wavelength of light used.

Solution:

26
2.8 Numerical problems - Interference by division of amplitude

Here, m + p = 15, m = 5. Therefore, p = 15 − 5 = 10.


And, D15 = 0.590 cm, D5 = 0.336 cm, R = 100 cm.
Therefore,
2 2
Dm+p − Dm 0.5902 − 0.3362
λ= =
4pR 4 × 10 × 100
0.3481 − 0.1129
=
4 × 103
= 0.0588 × 10−3 cm
= 5880 × 10−8 cm
= 5880 Å

Exercise 2.9
In a Newton’s rings formation in a liquid film with light of wavelength 6000
Å, diameter of the sixth bright ring is 3.1 mm and radius of curvature of the
curved surface is 1 m. Calculate the refractive index of the liquid.

Solution:

Here n = 6, D6 = 3.1 mm = 0.31 cm, R = 1 m = 100 cm, λ = 6000


Å= 6 × 10−5 cm.
For bright rings,
λR
Dn2 = 2(2n − 1)
µ
2(2n − 1)λR
=⇒ µ =
Dn2
Therefore, in this case,

2 × (12 − 1) × 6 × 10−5 × 1
µ=
0.31
2 × 11 × 6 × 10−5
=
0.31
≈ 1.374

27
2.9 Michelson interferometer

2.9 Michelson interferometer


The Michelson interferometer produces interference fringes by splitting a light
beam into two parts and then recombining them after they have travelled dif-
ferent optical paths. It has numerous applications, including precise length
measurement.
As shown in Figure 2.13, light from a diffuse light source S is made incident
on a half-silvered mirror M which passes half of the light and reflects the other
half to the mirror M1 . The passed (refracted) light is then made to pass through
a fully transparent compensation plate C onto a second mirror M2 . The com-
pensation plate is made from same material and has the same thickness as M.
This ensures both beams pass through the same thickness of glass, and any
phase difference between the two beams is only due to the difference in the
distances they travel.

Figure 2.13: Michelson interferometer (a) Typical experimental setup (b) Pla-
nar view showing the propagation of light in the set up (Source:
Physics LibreTexts).

The observer sees the interference pattern due to beams reflected from the
mirror M1 and a virtual image M02 of the mirror M2 formed by M, which act
as coherent sources. The arrangement is optically equivalent to a thin air film
between M1 and M02 .
The schematics in Figure 2.13 only a single reference light beam emergent
from the light source S. However, it is an extended source (rather than a point
source), which means there are light beams are emergent from various points
on the source and are incident on M at different angles. With monochromatic
light, the resultant of interference between all these beams from the extended
source is a circular pattern of alternating dark and bright fringes (similar in
appearance to Newton’s rings) when the two mirrors M1 and M02 are parallel

28
2.9 Michelson interferometer

to each other. Each circular ring corresponds to a certain inclination of the


incident beam with respect to the plane of the mirrors. Hence the fringes are
known as fringes of equal inclination. When M1 and M02 are not parallel, they
enclose a wedge-shaped film, and curved fringes can be observed. These are
also known as fringes of equal thickness. When M1 and M02 intersect, straight
line fringes are obtained around the point of intersection.

2.9.1 Suggested further reading: Michelson interferometer


1. Physics LibreTexts

2. Physical Optics Blog

3. Michelson interferometer@Physics StackExchange

4. Carl Wieman Science Education Initiative (CWSEI)

5. Michelson Interferometer Experiment manual@NISER

...and a lot more on the World Wide Web.

********************* The End (of this Unit) *********************

29

You might also like