Diffraction

PH-1007 (Physics)
Dr. Gorky Shaw

3.1 Introduction
According to geometrical optics, if a plane wave is incident on a long narrow
slit, as shown in Figure 3.1, the region PQ is illuminated and the rest of the
region forms the geometrical shadow.
However, if the slit is made very narrow (comparable to the wavelength of
light), then light bends into the geometrical shadow region. This is known as
diffraction.

Figure 3.1: Light bent into geometrical shadow by diffraction.

1. Diffraction is the slight bending of light into the geometrical shadow

region, when passing through a narrow slit, comparable to the wavelength
of light.
2. The resultant intensity distribution is known as diffraction pattern.
3. Diffraction can be explained based on Huygens principle which states that
each point on an wavefront acts as a source of secondary waves, called
wavelets. These superpose with each other to result in the phenomenon
of diffraction.
4. Diffraction results from the interference between waves from the same
source.
 3.2 Fraunhofer diffraction at a single slit
Some differences between interference and diffraction are:

1. Interference results from waves emerging from different sources. Diffrac-

tion results from waves emerging from the same source.

2. Diffraction maxima are never of the same width. Interference maxima

may or may not be of the same width.

3. Diffraction requires an obstacle in the path of light. Interference can occur

without any obstacle.

4. Direction propagation remains unaltered in interference, whereas light

gets bent and direction of propagation is modified in diffraction.

3.1.1 Types of diffraction

Diffraction can be categorized into two types:

1. Fresnel diffraction: The source, slit and screen are at finite distance
from each other. The incident wavefront is spherical or cylindrical.

2. Fraunhofer diffraction: The source and screen are effectively at infinite

distance from the diffracting obstacle. This is achieved by placing the
source and the screen at the focal planes of two lenses. The incident
wavefront is plane.

3.2 Fraunhofer diffraction at a single slit

3.2.1 Resultant of n simple harmonic waves
For n simple harmonic waves having equal amplitude a and common phase
difference d between successive vibrations, the resultant amplitude R is given
by
a sin(nd/2)
R= (3.1)
sin(d/2)

3.2.2 Resultant intensity due to Fraunhofer diffraction at a single slit

We consider Fraunhofer diffraction from a single slit AB as shown in Figure
3.2. According to Huygens principle, each point on slit AB sends out sec-
ondary wavelets in all directions. Rays which do not diffract focus at O. Those
diffracted through an angle θ are focused at P. We have, path difference between

2
 3.2 Fraunhofer diffraction at a single slit

Figure 3.2: Fraunhofer diffraction at a single slit.

wavelets reaching P from A and B,

BK = AB sin θ = e sin θ (3.2)

Corresponding phase difference

2π
= e sin θ (3.3)
λ
Let the width of AB be divided into n equal parts, and amplitude of wavelets
from each part be a.
Then, the phase difference between waves from consecutive parts is
 
1 2π
d= e sin θ (3.4)
n λ

Therefore, the resultant amplitude at P,

a sin(nd/2)
R=
sin(d/2)
a sin(πe sin θ/λ)
= (3.5)
sin(πe sin θ/nλ)

3
 3.2 Fraunhofer diffraction at a single slit

πe sin θ
Let = α. Then
λ
a sin(nd/2)
R=
sin(d/2)
a sin α
=
sin(α/n)
a sin α
' (for small α/n)
α/n
na sin α
= (3.6)
α
As n → ∞, a → 0. But the product na remains finite.
Let na = A. Then
A sin α
R= (3.7)
α
Therefore, resultant intensity at P,

A2 sin2 α
2
I=R = (3.8)
α2

3.2.3 Conditions for maxima and minima

For maximum or minimum intensity, we must have
dI
=0
dα
d A2 sin2 α
 
=⇒ =0
dα α2
2 sin2 α
 
2 2 sin α cos α
=⇒ A − =0
α2 α3
=⇒ sin α(α cos α − sin α) = 0 (3.9)

This means, either

sin α = 0 (3.10)

4
 3.2 Fraunhofer diffraction at a single slit

or

α cos α − sin α = 0
or, α − tan α = 0 (3.11)

3.2.4 Condition for minima

From (3.8) and (3.10), the intensity at P is minimum (Imin = 0) if

sin α = 0 (but α 6= 0)
=⇒ α = nπ (n = ±1, ±2, ±3...but n 6= 0)
πe sin θ
=⇒ = nπ (3.12)
λ
That is,

e sin θ = nλ (3.13)

First, second, third... minima occur at n = ±1, ±2, ±3... etc.

3.2.5 Principal maximum

sin α
When α = 0, the term becomes indeterminate. Clearly, this is not a
α
condition for minimum intensity. So, this has to be condition for maximum
intensity. To establish this, we consider the limit α → 0 and apply L’Hospital’s
rule as follows:
sin α cos α
lim = lim = 1 6= 0 (3.14)
α→0 α α→0 1
So, α = 0 does correspond to a maximum. Thus, we have

πe sin θ/λ = 0
=⇒ sin θ = 0
=⇒ θ = 0 (3.15)

That is the same direction as that of the incident light. This is known as the
principal maximum or central maximum.

5
 3.2 Fraunhofer diffraction at a single slit

3.2.6 Secondary maxima

Since there are a number of minima (given by the condition (3.13), there must
also be maxima in between any two of these minima). These are known as
secondary maxima, and the condition for these is given by (3.11), that is

α − tan α = 0 (3.16)

(3.16) can be solved graphically by plotting y = α and y = tan α and finding

the points of intersection. We can thus obtain the solutions of (3.16):
3π 5π 7π
α'± , ± , ± , ... (3.17)
2 2 2
apart, of course, from α = 0, which is the condition for the central maximum.
From (3.17), we have for secondary maxima,

πe sin θ π
' (2n + 1) , n = ±1, ±2, ±3... (3.18)
λ 2
or,

λ
e sin θ ' (2n + 1) , n = ±1, ±2, ±3... (3.19)
2

3.2.7 Ratio of intensities at the principal and secondary maxima

Intensity at the central maximum,
 2
sin 0
I0 = A2 = A2 (3.20)
0

At the secondary maxima,

2
4A2

sin(3π/2)
I1 = A2 = (for n = 1)
3π/2 9π 2
2
4A2

2 sin(5π/2)
I2 = A = (for n = 2)
5π/2 25π 2
2
4A2

2 sin(7π/2)
I3 = A = (for n = 3) (3.21)
7π/2 49π 2

6
 3.2 Fraunhofer diffraction at a single slit

etc.

So, intensities at successive maxima are in the ratio

4A2 4A2 4A2

1: : : ... (3.22)
9π 2 25π 2 49π 2
as represented schematically in Figure 3.3.

Figure 3.3: Intensity distribution due to Fraunhofer diffraction at a single slit.

7
 3.3 Plane Diffraction Grating

3.3 Plane Diffraction Grating

A plane diffraction grating is an arrangement having a large number of
parallel slits of the same width, separated by opaque spaces of equal width.

3.3.1 Resultant intensity

Let AB be the section of a grating with slit width e and the width of opaque
spaces d (as shown in Figure 3.4). The quantity (e + d) is called the Grating
element. Let N be the number of slits in the grating. Path difference between
rays from slits S1 and S2 at P ,

S2 K = S1 S2 sin θ = (e + d) sin θ (3.23)

Corresponding phase difference,

2π
= (e + d) sin θ = 2β (say) (3.24)
λ
where λ is wavelength of the incident monochromatic light.
Then, in direct analogy with the case of diffraction from a single slit, the
resultant amplitude at P is
0
A sin N β
R= (3.25)
sin β

0
where A = Resultant amplitude from each slit
A sin α
= (3.26)
α
That is,

A sin α sin N β
R= · (3.27)
α sin β

So, resultant intensity at P ,

0
A 2 sin2 N β
2 A2 sin2 α sin2 N β
I=R = = · (3.28)
sin2 β α2 sin2 β

8
 3.3 Plane Diffraction Grating

The first factor gives a diffraction pattern due to a single slit. The second
factor gives the interference pattern due to N slits.

Figure 3.4: Schematic representation of a plane diffraction grating.

3.3.2 Maxima and minima in the diffraction pattern

For maximum or minimum value of intensity I given in (3.4), we have the
condition
dI
=0
dβ
 0
d A 2 sin2 N β

=⇒ =0
dβ sin2 β
2N sin N β cos N β 2 sin2 N β cos β
=⇒ − =0
sin2 β sin3 β
=⇒ sin N β(N cos N β sin β − sin N β cos β) = 0

The above implies either

9
 3.3 Plane Diffraction Grating

sin N β = 0 (3.29)

or

N cos N β sin β − sin N β cos β = 0

Dividing by cos N β cos β and rearranging, we get

N tan β = tan N β (3.30)

3.3.3 Minima
When sin N β = 0 (but sin β 6= 0), we have the minimum intensity I = Imin = 0.
sin N β
When sin β = 0, the term is indeterminate (and corresponds to principal
sin β
maxima, as discussed in Section 3.3.4 below.
Therefore, the condition for minimum intensity, given by (3.29) is

sin N β = 0( but sin β 6= 0)

=⇒ N β = mπ (3.31)

π
=⇒ N (e + d) sin θ = mπ
λ

=⇒ N (e + d) sin θ = mλ (3.32)

where m takes all integral values except 0, ±N, ±2N , ±3N , ..., ±nN (where
n is an integer). This is because these values of m give sin β = 0, corresponding
to principal maxima, as discussed in Section 3.3.4. Thus, there are (N − 1)
minima between successive principal maxima.

10
 3.3 Plane Diffraction Grating

3.3.4 Principal maxima

sin N β
For maximum intensity, should be maximum. As discussed above, this
sin β
expression is indeterminate, and the intensity is maximum for for sin β = 0, or,

β = nπ, n = 0, ±1, ±2, ±3... (3.33)

Therefore we take limits and apply L’Hospitals’ rule:

sin N β N cos N β
lim = lim =N (3.34)
β→nπ sin β β→nπ cos β
Corresponding intensity,

A2 sin2 α
Ip = 2
· N2 (3.35)
α

These maxima are most intense and are called principal maxima. These
are obtained in the directions given by the condition (3.33):

β = nπ
π
=⇒ (e + d) sin θ = nπ
λ

=⇒ (e + d) sin θ = nλ (3.36)

where n = 0, ±1, ±2, ±3... correspond to the nth order principal maxima.

3.3.5 Secondary maxima

The (N −1) minima between each principal maximum produce (N −2) maxima,
called secondary maxima. These can be obtained from the condition (3.30),
sin N β
i.e., N tan β = tan N β. To find the corresponding values of , we use the
sin β
triangle as shown in Figure 3.5 which depicts the condition (3.30).

11
 3.3 Plane Diffraction Grating

sin N β
Figure 3.5: Triangle to determine .
sin β

From the triangle, we can see that

N tan β
sin N β = p (3.37)
1 + N 2 tan2 β
Therefore,

sin2 N β N 2 tan2 β
=
sin2 β (1 + N 2 tan2 β) sin2 β
N2
=
(cot2 β + N 2 ) sin2 β
N2
=
cos2 β + N 2 sin2 β
N2
= 2 2 (using sin2 β + cos2 β = 1) (3.38)
1 + (N − 1) sin β
Therefore, intensity of the secondary maxima,

0 A2 sin2 α N2
I = ·
α2 1 + (N 2 − 1) sin2 β

0 Ip
=⇒ I = (3.39)
1 + (N 2 − 1) sin2 β

The intensity of secondary maxima relative to principal maxima decreases

with increasing N .

12
 3.3 Plane Diffraction Grating

sin2 α sin2 N β
Variation of the and terms with the direction θ, and the
α2 sin2 β
resultant grating spectrum (intensity curve), are shown in Figure 3.6.

Figure 3.6: Grating spectrum

3.3.6 Missing orders or absent spectra

Missing orders or absent spectra occur occur when a principal maximum in
the grating spectrum overlaps with a minimum in the corresponding single slit
diffraction pattern.
From (3.36), principal maximum in the diffraction spectrum occur when

(e + d) sin θ = nλ, n = 0, ±1, ±2, ±3... (3.40)

Now, minima in a single slit pattern occur when

e sin θ = mλ, m = ±1, ±2, ±3... (3.41)

If both (3.40) and (3.41) are satisfied for a given θ, a particular maximum of
order n will be missing in the grating spectrum.

13
 3.4 Dispersive power and resolving power of a grating

Comparing (3.40) and (3.41), we have,

e+d n
= (3.42)
e m
Thus, for the nth order spectrum to be absent, for d = e,

n = 2m. i.e., n = 2, 4, 6... for m = 1, 2, 3... (3.43)

3.4 Dispersive power and resolving power of a grating

3.4.1 Dispersive power
The dispersive power of a grating
 is defined as the rate of variation of angle
dθ
of diffraction with wavelength .
dλ
Differentiating the condition for principal maxima for a given wavelength
with respect to λ, we get
dθ
(e + d) cos θ =n
dλ

dθ n
=⇒ = (3.44)
dλ (e + d) cos θ

3.4.2 Resolving power of an optical instrument: Rayleigh’s criterion of

resolution
Two images are just resolved if the position of the principal maximum in the
diffraction pattern of one coincides with the first minimum in the diffraction
pattern of the other. The criterion is illustrated in Figure 3.7. In panel (a), the
two neighbouring principal maxima corresponding to two wavelengths λ1 and
λ2 are clearly distinguishable due to the fairly wide region of low intensity in
between them. In panel (b), the two principal maxima are too close to each
other to be distinguished, as the resultant intensity shows no drop in between
them, and shows a peak instead. It would be difficult to ascertain whether the
pattern corresponds to two lines or just a single line. In panel (c), the principal
maximum of one overlaps exactly with the first minimum of the other, and
there is a dip in intensity in between the two, so that they are just resolved.

14
 3.4 Dispersive power and resolving power of a grating

Figure 3.7: Diffraction spectrum of two wavelengths.

3.4.3 Resolving power of a grating

The resolving power of a diffraction grating is defined as the capacity to form
separate diffraction maxima of two wavelengths which are very close to each
other. It is given by (λ/dλ), where dλ is the difference between two wavelengths
and λ is their mean wavelength.

Figure 3.8: Diffraction of two wavelengths at a grating.

We consider a parallel beam of light consisting of two wavelengths λ and

λ + dλ as shown in Figure 3.8.
For the nth principal maximum due to λ in the direction θn (at the point P1 ),
we have

(e + d) sin θn = nλ, n = 0, ±1, ±2, ±3... (3.45)

Now, the general condition for minima is (3.32)

N (e + d) sin θ = mλ, m 6= 0, ±N, ±2N , ±3N , ..., ±nN . (3.46)

15
 3.4 Dispersive power and resolving power of a grating

Figure 3.9: Overlapping of two diffraction spectra.

The first minimum of λ adjacent to the nth principal maximum, along the
direction (θn + dθn ), will correspond to m = nN + 1, for which we get

N (e + d) sin(θn + dθn ) = (nN + 1)λ (3.47)

If the nth principal maximum of λ + dλ, occurs in the same direction (see
Figure 3.9), then

(e + d) sin(θn + dθn ) = n(λ + dλ) (3.48)

Comparing (3.47) and (3.48), we get

(nN + 1)λ = nN (λ + dλ)

=⇒ nN λ + λ = nN λ + nN dλ
λ
=⇒ = nN (3.49)
dλ

Therefore,
λ
= nN
dλ
λ N (e + d) sin θn
or, = (3.50)
dλ λ

Thus, the resolving power of a grating is directly proportional to (i) Number

of lines (slits) in the grating, and (ii) Order of the spectrum.

16
 3.5 Numerical problems - Diffraction

3.5 Numerical problems - Diffraction

Exercise 3.1
Light of wavelength 5000 Å is incident normally on a plane transmission grat-
ing. Find the difference in deviations in the first and third order spectra. No.
of lines per cm on the grating surface is 6000.

Solution:

Given, λ = 5000 Å = 5×10−5 cm.

1
= 6000/cm.
e+d
First and third order spectra, correspond to n1 = 1 and n2 = 3, respectively.

For first order spectrum,

(e + d) sin θ1 = n1 λ
n1 λ 1 × 5 × 10−5
=⇒ sin θ1 = = = 0.3
e+d 1/6000

Therefore, θ1 ' 17.460 .

For third order spectrum,

(e + d) sin θ2 = n2 λ
n2 λ 3 × 5 × 10−5
=⇒ sin θ2 = = = 0.9
e+d 1/6000

Therefore, θ2 ' 64.160 .

Therefore, difference in deviation between the two spectra = θ2 − θ1 = 46.70

Exercise 3.2
Light of wavelength 5000 Å falls normally on a plane transmission grating
having 15000 lines in 3 cm. Find the angle of diffraction for maximum intensity

17
 3.5 Numerical problems - Diffraction

in the first order.

Solution:

Given, λ = 5000 Å = 5×10−5 cm.

3 1
e+d= = cm. For first order, n = 1.
15000 5000
Therefore,

nλ 1 × 5 × 10−5
sin θ = = = 0.25
e+d 1/5000

Therefore, θ ' 14.480 .

Exercise 3.3
A single slit is illuminated by light composed of two wavelengths λ1 and λ2 . It
is observed that the first diffraction minimum for λ1 coincides with the second
diffraction minimum for λ2 . What is the relation between λ1 and λ2 ?

Solution:

For minima of single slit diffraction intensity,

e sin θ = mλ

Given, minima corresponding to m = 1 for λ1 , and m = 2 for λ2 coincide, i.e.,

occur at the same θ.
Therefore,

e sin θ = λ1 = 2λ2
=⇒ λ1 = 2λ2

Exercise 3.4
What is the minimum number of lines required in grating to just resolve the
lines of wavelength 5890 Å and 5896 Å in the second order?

Solution:

18
 3.5 Numerical problems - Diffraction

Given, λ1 = 5890 Å, λ2 = 5896 Å, n = 2.

Therefore, dλ = 6 Å and mean wavelength λ = 5893 Å.

If N is the number of lines, then
λ
nN =
dλ
λ
=⇒ N =
ndλ
×5893
= ' 491
2×6

Exercise 3.5
How many orders will be observed by a grating with 4000 lines/cm, if it is
illuminated by light of wavelength in the range 5000-7500 Å?

Solution:
For principal maxima in a grating diffraction pattern,

nλ = (e + d) sin θ

Maximum order number observable with a grating corresponds to sin θ = 1,

that is,
e+d
nmax =
λ
1
Given, (e + d) = , λ1 = 5000 Å = 5×10−5 cm, λ2 = 7500 Å = 7.5×10−5
cm. 4000
So,
e+d 1
n1max = = =5
λ1 4000 × 5 × 10−5
and
e+d 1
n2max = = = 3.3
λ2 4000 × 7.5 × 10−5
n can only take integral values, so n2max = 3.
Therefore, 3 to 5 orders will be observed for the given wavelength range.

19