DOPPLER EFFECT IN LIGHT

174 Optics
!
y y

PN
R

a q0
a2 q2

q2 P4
a1 P L Nq0/2

(q 0 /2)
q1 q

)
sin (Nq
0 /2
O
x

a si n
q0
q0

C
P3
a

q0

P2
Fig. 13.8 The graphical method for determining the resultant
Q2
of two simple harmonic motions along the same q0
direction and having the same frequency. O Q1 P1 x
1 (N – 1)q
2 0

and a2, vector OR rotates on the circumference of the circle Fig. 13.9 The graphical method for determining the resultant
of radius OR with the same frequency. of N simple harmonic motions along the same
direction and having the same frequency.
Thus, if we wish to find the resultant of the two displace-
ments given by Eq. (20), then we must first draw a vector
( OP ) of length a1 making an angle q1 with the axis; from the
tip of this vector we must draw another vector ( PR ) of
length a2 making an angle q2 with the axis. The length of
vector OR will represent the resultant amplitude, and the In Fig. 13.9 vectors OP1 , P1 P2 , P2 P3 , . . . correspond to x1,
angle that it makes with the axis will represent the initial x2, x3, . . . , respectively. The resultant is denoted by vector
phase of the resultant displacement. It can be easily seen OPN . Let Q1L and Q2L be the perpendicular bisectors of
that if we have a third displacement OP1 and P1P2. It is easy to prove that
x 3 = a3 cos (wt + q3) (37) DLQ1P1 ∫ DLQ2P1
then from point R we must draw a vector RR¢ of length a3 Thus LO = LP1 = LP2. Therefore, points O, P1, P2, P3, . . ., PN
which makes an angle q3 with the axis; vector OR¢ will rep- will lie on the circumference of a circle whose center is L and
resent the resultant of x1, x2, and x3. whose radius is LO. Further, –LP1O = p – q0 / 2, and therefore
As an illustration of the above procedure, we consider the –OLP1 = q0. Thus,
resultant of N simple harmonic motions all having the same
amplitude and with their phases increasing in arithmetic pro- a /2
LO =
gression. Thus sin (q0 /2)
x 1 = a cos wt
Nq 0
x 2 = a cos (wt + q0) (38) and OPN = 2OC = 2LO sin
2
... ... ... ... ... ... ... ... ... sin Nq0 / 2
=a (39)
xN = a cos [wt + (N – 1)q0] sin q0 / 2

gha80482_ch13_167-176.PMD 174 1/16/2009, 5:55 PM

 304 WAVES AND OSCILLATIONS

In the case of transmitted rays there is only internal reflection at B and this causes no
additional change of phase. Thus, for the transmitted rays we have the following conditions
for maxima and minima:
maxima 2nd cos r = mλ (12.11a)
FG m + 1 IJ λ
minima 2nd cos r =
H 2K (12.11b)

with m = 0, 1, 2,...
These conditions are just opposite to those obtained with the reflected rays. As a result
a reflected and transmitted interference patterns are complementary to each other. This
means that the position which corresponds to the maximum of the reflected system will be
the position of minimum of the transmitted system and vice versa.
23. A water film (n = 4/3) in air is 315 nm thick. If it is illuminated with white light
at normal incidence, what colour will it appear to be in the reflected light?
Solution
2nd
For maxima, we have the relation λ =
dm + i 1
2

When m = 0, λ = 1680 nm (infrared),

m = 1, λ = 560 nm (visible),
m = 2, λ = 336 nm (ultraviolet).
Only the maximum corresponding to m = 1 lies in the visible region. Light of wavelength
560 nm appears yellow-green. The water film will appear yellow-green in the reflected light.
24. A lens is coated with a thin film of transparent substance magnesium fluoride
(MgF2) with n = 1.38 to reduce the reflection from the glass surface (n = 1.50). How thick a
coating is needed to produce a minimum reflection at the centre of the visible spectrum
(λ = 550 nm)?
Solution
We assume that the light strikes the lens at near-normal incidence. We would like to
find the thickness of the film that will bring about destructive interference between rays R1
and R2 (Fig. 12.9). Phase change of π is now associated with each ray and the condition of
minimum intensity is now
R1 R2

Air
n = 1.00

MgF2
n = 1.38

Glass
n = 1.50
Fig. 12.9
 INTERFERENCE 305

FG m + 1 IJ λ , m = 0, 1, 2,...
2nd =
H 2K
Thus, the thickness of the thin nest possible film is (for m = 0)
λ 550
d = = nm = 99.64 nm.
4 n 4 × 1.38
25. A soap film of thickness 5.5 × 10–5 cm is viewed at an angle of 45°. Its index of
refraction is 1.33. Find the wavelengths of light in the visible spectrum which will be absent
from the reflected light.
Solution
sin i
Since n = , cos r = 0.84696,
sin r
where i is the angle of incidence and r is the angle of refraction.
For minima, 2nd cos r = mλ
When m = 1, λ = 12.39 × 10–5 cm (infrared)
m = 2, λ = 6.20 × 10–5 cm (visible)
m = 3, λ = 4.13 × 10–5 cm (visible)
m = 4, λ = 3.10 × 10–5 cm (ultra violet)
Hence the absent wavelengths in the reflected light are 6.20 × 10–5 cm and
4.13×10–5 cm in the visible region of spectrum.
26. A tanker leaks kerosene (n = 1.2) into the Persian Gulf, creating a large slick on top
of the water (n = 1.3). (a) If you are looking straight down from an airplane onto a region of
the slick where its thickness is 460 nm, for which wavelength(s) of visible light is the reflection
the greatest? (b) If you are scuba-diving directly under this same region of the slick, for which
wavelength(s) of visible light is the transmitted intensity the strongest.
Solution
(a) For the reflected rays, the condition for maxima is [see problem 24]
2nd = mλ,
2nd 2 × 1.2 × 460
or λ = = nm
m m
The wavelength of visible light (for m = 2) is 552 nm.
(b) For the transmitted rays, the condition for maxima is

FG m + 1 IJ λ
2nd =
H 2K
2nd 2 × 1.2 × 460
or λ = = nm.
1 1
m+ m+
2 2
The wavelengths of visible light are 736 nm (when m = 1) and 441.6 nm
(when m = 2). [For the transmitted rays an extra phase change of π is introduced in the ray
AFBCG whereas no change of phase occurs in the ray AFH (Fig. 12.8) under the conditions
stated in (b)].
 306 WAVES AND OSCILLATIONS

27. A plane wave of monochromatic light falls normally on a uniform thin film of oil
that covers a glass plate. The wavelength of the source can be varied continuously. Complete
destructive interference of the reflected light is observed for wavelengths of 500 and 700 nm
and for no wavelengths between them. If the index of refraction of the oil is 1.3 and that of
glass is 1.5, find the thickness of the oil film.
Solution
The condition of destructive interference of the reflected light is (see problem 24)
2nd = (2m + 1)λ/2
or 4nd = (2m + 1)λ
Thus, 4 × 1.3 × d = 700 (2m + 1) = 500[2(m + 1) + 1]
which gives m = 2 and d = 673.08 nm.
28. Find the conditions of formation of bright and dark fringes produced by a wedge-
shaped thin film for near normal incidence.
Solution
We consider two plane surfaces OA and OB B
inclined at an angle θ which is very small
(Fig. 12.10). The thickness of the film increases P
from O to A. When the film is viewed with n
t
reflected monochromatic light, a system of
q
equidistant interference fringes are observed O
A
which are parallel to the line of intersection of
x
the two surfaces. A parallel beam of x1
monochromatic light falls on the face OB almost Fig. 12.10
normally. The components reflected from the
upper and the lower surfaces will interfere to produce alternatively dark and bright bands.
Since the angle of wedge θ is very small, the beam incident normally to the face OB may very
well be considered normal to the face OA too. The path difference between the two reflected
beams is (2nt + λ/2) at the point P where the thickness of the wedge is t. So at this position
of the film there will be dark fringe when
λ FG 1
= m + λ,
IJ
2nt +
2 H 2 K m = 0, 1, 2,....

or 2nt = mλ (minima)
where n is the refractive index of the film. For bright fringes, we have
FG m + 1 IJ λ (maxima)
2nt =
H 2K
We know θ = t/x or t = xθ (see Fig. 12.10). Thus mth dark fringe appears at P when
2nxθ = mλ
Similarly, if (m + p)th dark fringe is formed at a distance x1 from O, we have
2nx1θ = (m + p)λ
Thus, 2n(x1 – x)θ = pλ
We can count the number of fringes in a space (x1 – x) with a travelling microscope.
Here x1 – x is the distance corresponding to p fringes. The fringe width β is given by
x1 − x λ
β = = .
P 2nθ
 INTERFERENCE 307

If air is enclosed between OA and OB, n = 1, and we have

λ
β = .
2θ
29. Two glass plates enclose a wedge-shaped air film, touching at one edge and are
separated by a wire of 0.04 mm diameter at a distance of 10 cm from the edge. Calculate the
fringe width. Monochromatic light of wavelength 589.3 nm from a broad source falls normally
on the film.
Solution
Referring to the figure 12.10, we have OA = 10 cm and AB = 0.04 mm so that
AB 0.004
θ = = = 0.0004 rad.
OA 10
λ 589.3 nm
Fringe width β = = = 0.074 cm.
2θ 2 × 0.0004
30. A broad source of light (λ = 680 nm) illuminates normally two glass plates 120 mm
long that touch at one end and are separated by a wire 0.048 mm in diameter at the other
end. How many bright fringes appear over the 120 mm distance?
Solution
λ 680 × 120
Fringe width β = = nm
2θ 2 × 0.048
Number of bright fringes over the 120 mm distance
120 × 10 6
= = 141.
β
31. Two wires with diameters 0.01 cm and 0.03 cm are laid parallel to each other and
2 cm apart on a piece of plane glass. Another piece of plane glass is laid on top. Monochromatic
light of wavelength 546 nm falls normally on the glass plates. Calculate the distance between
consecutive bright fringes.
Solution
0.03 − 0.01
Angle of the wedge θ = = 0.01 rad.
2
λ 546
Fringe width β = = nm = 27.3 µm.
2θ 2 × 0.01
32. In an air wedge formed by two plane glass plates touching each other along one edge
there are 4001 dark lines observed when viewed by reflected monochromatic light. When the
air between the plates is evacuated, only 4000 such lines are observed. Calculate the index of
refraction of the air from this data.
Solution
If mth dark fringe appears at a distance x from the edge then
x = mλ/(2nθ)
In the present problem we have
4001λ 4000λ
x = =
2nθ 2θ
4001
which gives n = = 1.00025.
4000
20 Simple Harmonic Motion

Superposition of a Large Number n of Simple Harmonic Vibrations

of Equal Amplitude a and Equal Successive Phase Difference d
Figure 1.11 shows the addition of n vectors of equal length a, each representing a simple
harmonic vibration with a constant phase difference ! from its neighbour. Two general
physical situations are characterized by such a superposition. The first is met in Chapter 5
as a wave group problem where the phase difference ! arises from a small frequency
difference, !!, between consecutive components. The second appears in Chapter 12 where
the intensity of optical interference and diffraction patterns are considered. There, the
superposed harmonic vibrations will have the same frequency but each component will have
a constant phase difference from its neighbour because of the extra distance it has travelled.
The figure displays the mathematical expression

R cos ð!t þ "Þ ¼ a cos !t þ a cos ð!t þ !Þ þ a cos ð!t þ 2!Þ

þ % % % þ a cos ð!t þ ½n ' 1(!Þ

C
r
a δ
O
nδ

δ
a δ

δ
n
n 2
s i a
r r 2r δ
=
R

a
δ
a δ
a = 2r sin
90° −
nδ 90° − δ 2
2 2 δ
a
α δ
A
a B

Figure 1.11 Vector superposition of a large number n of simple harmonic vibrations of equal
amplitude a and equal successive phase difference !. The amplitude of the resultant
n! sin n!=2
R ¼ 2r sin ¼a
2 sin !=2
and its phase with respect to the first contribution is given by
" ¼ ðn ' 1Þ!=2
Superposition of a Large Number n of Simple Harmonic Vibrations 21

where R is the magnitude of the resultant and " is its phase difference with respect to the
first component a cos !t.
Geometrically we see that each length
!
a ¼ 2r sin
2
where r is the radius of the circle enclosing the (incomplete) polygon.
From the isosceles triangle OAC the magnitude of the resultant
n! sin n!=2
R ¼ 2r sin ¼a
2 sin !=2
and its phase angle is seen to be
^ B ' OA
" ¼ OA ^C

In the isosceles triangle OAC

^ AC ¼ 90 ) ' n!
O
2
and in the isosceles triangle OAB

^ B ¼ 90 ) ' !
OA
2
so
! " ! "
) ! ) n! !
" ¼ 90 ' ' 90 ' ¼ ðn ' 1Þ
2 2 2

that is, half the phase difference between the first and the last contributions. Hence the
resultant
# $
sin n!=2 !
R cos ð!t þ "Þ ¼ a cos !t þ ðn ' 1Þ
sin !=2 2

We shall obtain the same result later in this chapter as an example on the use of exponential
notation.
For the moment let us examine the behaviour of the magnitude of the resultant

sin n!=2
R¼a
sin !=2

which is not constant but depends on the value of !. When n is very large ! is very small
and the polygon becomes an arc of the circle centre O, of length na ¼ A, with R as the
chord. Then
! n!
" ¼ ðn ' 1Þ *
2 2
22 Simple Harmonic Motion

A sinα (c)
R=
α (b) R A

A=na 2A
π

π α
0 π π 3 2π
2 2

(e)
(d)

A A = 3 circumference
2

Figure 1.12 (a) Graph of A sin "=" versus ", showing the magnitude of the resultants for (b)
" ¼ 0; (c) " ¼ #/2; (d) " ¼ # and (e) " ¼ 3#/2

and
! ! "
sin ! *
2 2 n

Hence, in this limit,

sin n!=2 sin" sin " A sin "

R¼a ¼a ¼ na ¼
sin !=2 "=n " "

The behaviour of A sin "=" versus " is shown in Figure 1.12. The pattern is symmetric
about the value " ¼ 0 and is zero whenever sin " ¼ 0 except at " ! 0 that is, when sin
"=" ! 1. When " ¼ 0, ! ¼ 0 and the resultant of the n vectors is the straight line of length
A, Figure 1.12(b). As ! increases A becomes the arc of a circle until at " ¼ #=2 the first and
last contributions are out of phase ð2" ¼ #Þ and the arc A has become a semicircle of
which the diameter is the resultant R Figure 1.12(c). A further increase in ! increases " and
curls the constant length A into the circumference of a circle (" ¼ #) with a zero resultant,
Figure 1.12(d). At " ¼ 3#=2, Figure 1.12(e) the length A is now 3/2 times the
circumference of a circle whose diameter is the amplitude of the first minimum.

+ Superposition of n Equal SHM Vectors of Length a with

Random Phase
When the phase difference between the successive vectors of the last section may take
random values $ between zero and 2# (measured from the x axis) the vector superposition
and resultant R may be represented by Figure 1.13.

+ This section may be omitted at a first reading.

Superposition of n Equal SHM Vectors of Length a with Random Phase 23

x
pffiffiffi
Figure 1.13 The resultant R ¼ na of n vectors, each of length a, having random phase. This result
is important in optical incoherence and in energy loss from waves from random dissipation processes

The components of R on the x and y axes are given by

R x ¼ a cos $ 1 þ a cos $ 2 þ a cos $ 3 . . . a cos $ n

X n
¼a cos $ i
i¼1

and
n
X
Ry ¼ a sin $ i
i¼1

where

R 2 ¼ R 2x þ R 2y

Now
!2 2 3
n
X Xn n
X n
X
R 2x ¼ a 2 cos $ i ¼ a 24 cos 2 $ i þ cos $ i cos $ j 5
i¼1 i¼1 i¼1 j¼1
i6¼j

In the typical term 2 cos $ i cos $ j of the double summation, cos $ i and cos $ j have random
values between , 1 and the averaged sum of sets of these products is effectively zero.
The summation
n
X
cos 2 $ i ¼ n cos 2 $
i¼1
24 Simple Harmonic Motion

that is, the number of terms n times the average value cos 2 $ which is the integrated value
of cos 2 $ over the interval zero to 2# divided by the total interval 2#, or
ð 2#
1 1
cos 2 $¼ cos 2 $ d$ ¼ ¼ sin 2 $
2# 0 2

So
n
X na 2
R 2x ¼ a 2 cos 2 $ i ¼ na 2 cos 2 $ i ¼
i¼1
2

and
n
X na 2
R 2y ¼a 2
sin 2 $ i ¼ na 2 sin 2 $ i ¼
i¼1
2

giving

R 2 ¼ R 2x þ R 2y ¼ na 2

or
pffiffiffi
R¼ na

Thus, the amplitude R of a system subjected

pffiffiffi to n equal simple harmonic motions of
amplitude a with random phases in only na whereas, if the motions were all in phase R
would equal na.
Such a result illustrates a very important principle of random behaviour.

(Problem 1.17)

Applications
Incoherent Sources in Optics The result above is directly applicable to the problem of
coherence in optics. Light sources which are in phase are said to be coherent and this
condition is essential for producing optical interference effects experimentally. If the
amplitude of a light source is given by the quantity a its intensity is proportional to a 2 , n
coherent sources have a resulting amplitude na and a total intensity n 2 a 2. Incoherent
sources
pffiffiffi have random phases, n such sources each of amplitude a have a resulting amplitude
na and a total intensity of na 2 .

Random Processes and Energy Absorption From our present point of view the
importance of random behaviour is the contribution it makes to energy loss or absorption
from waves moving through a medium. We shall meet this in all the waves we discuss.
Some Useful Mathematics 25

Random processes, for example collisions between particles, in Brownian motion, are of
great significance in physics. Diffusion, viscosity or frictional resistance and thermal
conductivity are all the result of random collision processes. These energy dissipating
phenomena represent the transport of mass, momentum and energy, and change only in the
direction of increasing disorder. They are known as ‘thermodynamically irreversible’
processes and are associated with the increase of entropy. Heat, for example, can flow only
from a body at a higher temperature to one at a lower temperature. Using the earlier
analysis where the length a is no longer a simple harmonic amplitude but is now the
average distance a particle travels between random collisions (its mean free path), we see
that after n such collisions (with, on average, equal timepffiffiffiintervals between collisions) the
particle will, on average, have travelled only a distance na from its position at time t ¼ 0,
so that the distance travelled varies only with the square root of the time elapsed instead of
being directly proportional to it. This is a feature of all random processes.
pffiffiffi
Not all the particles of the system will have travelled a distance na but this distance is
the most probable and represents a statistical average.
Random behaviour is described by the diffusion equation (see the last section of
Chapter 7) and a constant coefficient called the diffusivity of the process will always
arise. The dimensions of a diffusivity are always length 2 /time and must be interpreted in
terms of a characteristic distance of the process which varies only with the square root of
time.

Some Useful Mathematics

The Exponential Series
By a ‘natural process’ of growth or decay we mean a process in which a quantity changes
by a constant fraction of itself in a given interval of space or time. A 5% per annum
compound interest represents a natural growth law; attenuation processes in physics usually
describe natural decay.
The law is expressed differentially as

dN dN
¼ ," dx or ¼ ," dt
N N

where N is the changing quantity, " is a constant and the positive and negative signs
represent growth and decay respectively. The derivatives dN/dx or dN/dt are therefore
proportional to the value of N at which the derivative is measured.
Integration yields N ¼ N 0 e ,"x or N ¼ N 0 e ,"t where N 0 is the value at x or t ¼ 0 and e
is the exponential or the base of natural logarithms. The exponential series is defined as

x2 x3 xn
ex ¼ 1 þ x þ þ þ %%% þ þ %%%
2! 3! n!

and is shown graphically for positive and negative x in Figure 1.14. It is important to note
that whatever the form of the index of the logarithmic base e, it is the power to which the