Lecture 2: Wavefront and Amplitude Division Class of

Interference

1 Wavefront division class of interference

In this class of interference, a single wavefront produces two waves of equal phase of which
superposition creates the interference pattern. Take a wavefront, which corresponds to the
locus of constant phase. Take any two part of the wavefront and allow them to superpose as
shown in figure 1. Both S1 and S2 have the same phase as they are on an iso-phase contour.

Figure 1: Interference by division of incident waveform.

So if S1 and S2 are sufficiently close, then spatial correlation will also exist. The above depicts
a possible ways to do that but there are a few experimental set up that does the same, which
we discuss next.

• Fresnel’s bi-prism: The first such setup corresponds to that of Fresnel’s bi-prism, which
has been depicted in figure 2. The bases of two prisms are joined together as shown in

1
Optics Indian Association for the Cultivation of Science

figure 2. So Rays from a source S diverges after refraction, so that they apparently
originate from S1 and S2 . Thus it appears as if S1 and S2 are interfering as separate
sources. The interference pattern is produced on the screen.

Figure 2: Fresnels’ bi-prism.

• Lloyd’s mirror: The second such set up is the Lloyd’s mirror (see figure 3). Light from
a single source S reflects on the mirror M and there are two apparent sources S and S 0 ,
which interfere. The interference pattern is formed on the screen.

Figure 3: Lloyd’s mirror.

• Fresnel’s double mirror: As a final example, we consider the Fresnel’s double mirror set
up (see figure 4). This involves two mirrors inclined at a certain angle among themselves.
As the rays reflect from these mirrors, they apparently appear from S1 and S2 . Thus we
have interference pattern on the screen.

2 Some useful applications of Interference

In this section we will demonstrate a few useful applications of the interference phenomenon.

2
Optics Indian Association for the Cultivation of Science

Figure 4: Fresnel’s double mirror.

• Finding wavelength of light: As we have seen in the previous lecture, the fringe width
of the interference pattern takes the following form
λD
β= (1)
d
Thus knowing distance between the screen and the source (D) and the distance between
the two sources (d), along with the information of the fringe width (β) we can easily
calculate the wavelength of the light source by the relation λ = βd/D.

• Finding the thickness of a thin film: Using interference one can determine the
thickness of a thin film. This is because the path difference will change as the thin film is
introduced (see figure 5). Suppose originally we had a bright fringe at point P, at distance
y above the central point C, in absence of any film. Thus the path difference between the
two light rays, in this context is given by
yn d
S2 P − S1 P = nλ = ; yn = nβ (2)
D
As we introduce a thin film in between the slit and the screen the effective path difference
between the light rays get changed and m bright fringes gets shifted in the screen. As a
result the equation for constructive interference will change as
yn−m d
S2 P − (S1 P − L) − µL =
D
yn−m d
=⇒ S2 P − S1 P − (µ − 1)L =
D
yn d yn−m d
=⇒ − = (µ − 1)L
D D
d
=⇒ (yn − yn−m ) = (µ − 1)L (3)
D

3
Optics Indian Association for the Cultivation of Science

Figure 5: Setup to determine thickness of thin film.

We know that, (d/D) = (λ/β), thus we obtain

λ
(yn − ym )= (µ − 1)L
β
(yn − ym )λ
=⇒ L = , (4)
(µ − 1)β

where the last equation determines the thickness of the film in terms of — (a) refractive
index µ, (b) wavelength of light λ, (c) fringe width β and (d) the separation due to
introduction of film yn − yn−m . All of these quantities, if known, the thickness of the film
can be determined.

• Angle of bi-prism The angle of Fresnel’s bi-prism introduced in the previous section
can also be determined using interference fringes. Using the formula for deflection angle
due to prism, we argue that each ray entering one prism gets deflected by

δ = (µ − 1)A (5)

where A is the angle of the prism. Thus, the separation between the apparent sources S1
and S2 is given by (see figure 2),

d = 2aδ
= 2a(µ − 1)A
d
=⇒ A = (6)
2a(µ − 1)

Thus knowledge about, d (which will come from interference), the refractive index µ and
the separation between source and the bi-prism a enables one to determine the angle of
the bi-prism.

4
Optics Indian Association for the Cultivation of Science

3 Amplitude Division class of Interference

If a wave of amplitude A falls on a dividing surface, then a part of it is reflected and a part
is transmitted. The ratio of the amplitude of reflected wave to the incident wave is known as
reflection coefficient r. Similarly, the ratio of the amplitude of transmitted wave and incident
wave is known as transmission coefficient t. Thus,
Amplitude of the reflected wave
Reflection coefficient (r) = (7)
Amplitude of the incident wave
and
Amplitude of the transmitted wave
Transmission coefficient (t) = (8)
Amplitude of the incident wave

The most common example, where such a class of interference becomes important, is when
multiple reflection and refractions are used to form interference pattern. Following which con-
sider a single beam of light ray falling onto a glass slab. So there will be multiple reflected and
transmitted ray (see figure 6). For example, a ray could be directly reflected, while another one
will be initially refracted, then reflected. There will be of course be path difference between
these two types of rays, so when converged using lens they will produce bright/dark fringes.

Figure 6: Amplitude division class of interference.

Path difference between two successive reflected rays: For this derivation we will
closely follow figure 6. Let the incident ray be P Q. From the top of the surface it gets reflected
along QR1 and refracted along QR. The refracted ray, at lower surface get reflected along RS.
This reflected ray RS is further refracted at the top of the surface along SR2 in the incident
medium. The superposition of the rays QR1 and SR2 will produce interference pattern. To see
various properties of this interference pattern, we would like to compute their path difference,
which is given by
∆ = µ(QR + RS) − QN , (9)

5
Optics Indian Association for the Cultivation of Science

where µ is the refractive index of the material considered here. From the figure it is obvious
that, RS = RN3 and hence QR + RS = QR + RN3 = QN3 . Thus the path difference becomes,
∆ = µ(QN3 ) − QN (10)
One can further use Snell’s law, which states sin φ1 = µ sin φ2 and thus from the geometry of
the setup we have,
QN QN̄
= sin φ = µ sin φ2 = µ
QS QS
=⇒ QN =µQN̄ (11)
Therefore, the path difference ∆ becomes
∆ = µ(QN3 ) − µ(QN̄ )
= µ(N̄ N3 )
= µ × SN3 × cos φ2
= µ × 2d × cos φ2
= 2µd cos φ2 , (12)
where, d is the thickness of the plate. This provides the path difference between two successive
rays. However there are additional subtleties associated with the above problem, which we
discuss below in the form of Stoke’s law.

Stokes’ Law: According to Stoke’s law, either the reflection from denser to rarer medium,
or the reflection from rarer to denser medium should have an additional phase shift by π.
Which one among the above two should have the phase shift, is determined by the Llyod’s
mirror experiment, where the central fringe has been observed to be dark. This suggests that
reflection from rarer to denser medium should inherit a phase shift of π.

In the present setup, as the ray P Q is reflected from denser to rarer medium, there will be an
additional path difference ±λ/2 which corresponds to the phase shift by an angle π. Thus, the
actual path difference is given by,
λ
∆actual = 2µd cos φ2 ± (13)
2
Therefore, for the bright fringe, we have the following condition
λ
2µd cos φ2 ± = nλ
2
 
1
=⇒ 2µd cos φ2 = n ∓ λ, (14)
2
where, n = 0, 1, 2, . . .. Similarly, the condition for appearance of a dark fringe is
λ λ
2µd cos φ2 ± = (2n ∓ 1)
2 2
=⇒ 2µd cos φ2 = nλ (15)
where, n = 0, 1, 2, . . ..

6
Optics Indian Association for the Cultivation of Science

Nature of fringes: As figure 7 demonstrates we can consider two co-linear sources of light
rays, which are coherent with the path difference as depicted above and located at (0, D+d/2, 0)
and (0, D − d/2, 0) respectively. If we consider the interference to be happening at a point on
the screen located at (x, 0, z), then the path difference will be

"  2 #1/2 "  2 #1/2

d d λ
Path Difference = x2 + z 2 + D + − x2 + z 2 + D − − (2n ± 1) (16)
2 2 2

If we want this point to have constructive interference the above path difference must be equal

Figure 7: Illustration of path difference computation.

to pλ, where p is an integer. Thus we must have,

"  2 #1/2 "  2 #1/2

d d λ
x2 + z 2 + D + − x2 + z 2 + D − = [2(n + p) ± 1]
2 2 | {z }2
m
 2  2  2 "  2 #1/2
d mλ d mλ d
=⇒ x2 + z 2 + D + = + x2 + z 2 + D − +2× × x2 + z 2 + D −
2 2 2 2 2
 2 "  2 #1/2
mλ 2 2 d
=⇒ 2Dd = + mλ x + z + D −
2 2
s 2
m2 λ2

2 2
d
=⇒ 2Dd − = mλ x + z + D −
4 2
2
m2 λ2
 
 2 2Dd −
d 4
=⇒ D − + x2 + z 2 =
2 m2 λ2
d2 4d2 m2 λ2
 
2 2
=⇒ x + z = D −1 − 2
+ + (17)
4D2 m2 λ2 16

7
Optics Indian Association for the Cultivation of Science

With the following two natural assumptions, (m2 λ2 /16D2 )  1 and (d2 /4D2 )  1, the above
equation can be expressed as,

4d2
 
2 2
x +z =D 2
− 1 ≡ Rm 2
(18)
m2 λ2

Thus the fringes are circular in nature, with the radius of concentric fringes being given by
s  
2Dd mλ mλ
Rm = 1+ 1− (19)
mλ 2d 2d
r
2Dd 1 λ2
= − (20)
λ m2 4d2
Thus as m (denoting the order of fringes) increases, the radius of the corresponding fringe
decreases.

Fringe width: The fringe width between mth and (m + 1)th order bright fringes can be
expressed as
2 2
Rm − Rm+1 R2 − Rm+1
2
βm = Rm − Rm+1 = ≈ m (21)
Rm + Rm+1 2Rm
Hence, using the expression for Rm , the fringe width takes the following expression,

4D2 d2 1
 
1
− −1/2
m2 m2 λ2
 
λ2 m2 (m + 1)2 2Dd
βm = 1/2 = 1− 1−
λ2 mλ (m + 1)2 4d2

2Dd 1
−
λ m2 4d2
−1/2
m2 λ2

4Dd
=⇒ βm ≈ 2 1− (22)
mλ 4d2

Thus given various parameters associated with the problem, one can immediately determine
the fringe width βm associated with mth circular fringe.