Lecture 6:

Waves Review, Crystallography,

and Examples

Lecture 6, p. 1
 Single-Slit Diffraction Example
Suppose that when we pass red light (λ = 600 nm) through a slit of
unknown width a, the width of the spot (the distance between the first
zeros on each side of the bright peak) is W = 1 cm on a screen that is
L = 2 m behind the slit. How wide is the slit?

a
θ
W = 1 cm

L=2m

Lecture 6, p. 2
 Solution
Suppose that when we pass red light (λ = 600 nm) through a slit of
unknown width a, the width of the spot (the distance between the first
zeros on each side of the bright peak) is W = 1 cm on a screen that is
L = 2 m behind the slit. How wide is the slit?

a
θ
W = 1 cm

L=2m

The angle to the first zero is: θ = ±λ/a

W = 2L tanθ ≅ 2Lθ = 2Lλ/a
a = 2Lλ/W = (4m)(6×10-7 m) /(10-2 m) = 2.4×10-4 m = 0.24 mm

Lecture 6, p. 3
 Multiple Slit Interference (from L4)
9I1 16I1 25I1
N=3 N=4 N=5
20

I3 I4 I5
10

0 φ 0 0
−2π 0 2π −2π 0 2π φ −2π 0 2π φ
−λ/d 0 λ/d θ −λ/d 0 λ/d θ −λ/d 0 λ/d θ

The positions of the principal maxima occur at φ = 0, ±2π, ±4π, ...

where φ is the phase between adjacent slits. θ = 0, ±λ/d, ±2λ/d, ...

The intensity at the peak of a principal maximum goes as N2.

3 slits: Atot = 3A1 ⇒ Itot = 9I1. N slits: IN = N2I1.

Between two principal maxima there are N-1 zeros and

N-2 secondary maxima ⇒ The peak width ∝ 1/N.

The total power in a principal maximum is proportional to N2(1/N) = N.

Lecture 6, p. 4
 Multiple-slit Example
y
Three narrow slits with equal spacing d are at
a distance L = 1.4 m away from a screen. The
slits are illuminated at normal incidence with
θ
light of wavelength λ = 570 nm. The first principal
d L
maximum on the screen is at y = 2.0 mm.

1. What is the slit spacing, d?

2. If the wavelength, λ, is increased, what happens to the width of the

principal maxima?

Lecture 6, p. 5
 Solution
y
Three narrow slits with equal spacing d are at
a distance L = 1.4 m away from a screen. The
slits are illuminated at normal incidence with
θ
light of wavelength λ = 570 nm. The first principal
d L
maximum on the screen is at y = 2.0 mm.

1. What is the slit spacing, d?

The first maximum occurs when the path difference between adjacent slits is
λ. This happens at sinθ = λ/d. We are told that tanθ = y/L = 1.43X10-3, so
the small angle approximation is OK. Therefore, d ≈ λ/θ = 0.40 mm.

2. If the wavelength, λ, is increased, what happens to the width of the

principal maxima?
The relation between θ and φ is φ/2π = δ/λ = d sinθ / λ.
Therefore, for every feature that is described by φ (peaks, φ
minima, etc.) sinθ is proportional to λ. The width increases.

Lecture 6, p. 6
 ACT 1: Multiple Slits
I =?
9I1
1. In 2-slit interference, the first minimum
corresponds to φ = π. For 3-slits, we have a
secondary maximum at φ = π (see diagram). I
What is the intensity of this secondary
maximum?
0
c) 3 I1 −2π 0 2π
a) I1 b) 1.5 I1 φ
φ=?
2. What value of φ corresponds to the first zero of the
3-slit interference pattern?
a) φ=π/2 b) φ=2π/3 c) φ=3π/4

3. What value of φ corresponds to the first zero of the

4-slit interference pattern?

a) φ=π/2 b) φ=2π/3 c) φ=3π/4

Lecture 6, p. 7
 Solution
I =?
9I1
1. In 2-slit interference, the first minimum
corresponds to φ = π. For 3-slits, we have a
secondary maximum at φ = π (see diagram). I
What is the intensity of this secondary
maximum?
0
c) 3 I1 −2π 0 2π
a) I1 b) 1.5 I1 φ
φ=?
Draw the phasor diagram :
A1
=
Two phasors cancel, leaving only one
I1

2. What value of φ corresponds to the first zero of the

3-slit interference pattern?
a) φ=π/2 b) φ=2π/3 c) φ=3π/4
Lecture 6, p. 8
 Solution
I =?
9I1
2. What value of φ corresponds to the first
zero of the 3-slit interference pattern?
I
a) φ=π/2 b) φ=2π/3 c) φ=3π/4

0
−2π 0 2π
φ
φ φ=?
φ
φ
φ=3π/4
A φ=π/2 φ=2π/3
A
No. Yes! No.
A is not zero. Equilateral triangle Triangle does not close.
gives A = 0.
3. What value of φ corresponds to the first zero of the
4-slit interference pattern?

a) φ=π/2 b) φ=2π/3 c) φ=3π/4

Lecture 6, p. 9
 Solution
I =?
9I1
3. What value of φ corresponds to the first zero
of the 4-slit interference pattern?
I

a) φ = π/2 b) φ = 2π/3 c) φ = 3π/4

0
−2π 0 2π
φ
φ=?
φ
φ
φ
φ = 2π/3 φ = 3π/4
φ = π/2 A
A

Yes. The square No. With 4 slits we

gives A = 0. No. A ≠ 0
“wrap around” and
A = A1.

To get a zero, we need a closed figure.

Nφ must be a multiple of 2π, so the first zero is at 2π/N.
Lecture 6, p. 10
 Interference & Diffraction Exercise Imax
Light of wavelength λ is incident on an N-slit system
with slit width a and slit spacing d.
I
1. The intensity I as a function of y at a viewing screen
located a distance L from the slits is shown to the right.
L >> d, y, a. What is N? 0
−6 0 +6
y (cm)
a) N = 2 b) N = 3 c) N = 4

2. Now the slit spacing d is halved, but the slit width a is kept constant.
Which of the graphs best represents the new intensity distribution?

a) Imax b) Imax c) Imax

I I I

0 0 0
−6 0 +6 −6 0 +6 −6 0 +6
y (cm) y (cm) y (cm)

Lecture 6, p. 11
 Solution Imax
Light of wavelength λ is incident on an N-slit system
with slit width a and slit spacing d.
I
1. The intensity I as a function of y at a viewing screen
located a distance L from the slits is shown to the right.
L >> d, y, a. What is N? 0
−6 0 +6
y (cm)
a) N = 2 b) N = 3 c) N = 4

N is determined from the number of minima between two principal maxima.

N = #minima+1 Therefore, N = 3 .
2. Now the slit spacing d is halved, but the slit width a is kept constant.
Which of the graphs best represents the new intensity distribution?

a) Imax b) Imax c) Imax

I I I

0 0 0
−6 0 +6 −6 0 +6 −6 0 +6
y (cm) y (cm) y (cm)

Lecture 6, p. 12
 Solution Imax
Light of wavelength λ is incident on an N-slit system
with slit width a and slit spacing d.
I
1. The intensity I as a function of y at a viewing screen
located a distance L from the slits is shown to the right.
L >> d, y, a. What is N? 0
−6 0 +6
y (cm)
a) N = 2 b) N = 3 c) N = 4

N is determined from the number of minima between two principal maxima.

N = #minima+1 Therefore, N = 3 .
2. Now the slit spacing d is halved, but the slit width a is kept constant.
Which of the graphs best represents the new intensity distribution?

a) Imax b) Imax c) Imax

I I I

0 0 0
−6 0 +6 −6 0 +6 −6 0 +6
y (cm) y (cm) y (cm)
Interference spacing Diffraction profile Interference spacing doubles.
should change. shouldn’t change. Diffraction profile is unchanged.
Lecture 6, p. 13
 Crystal diffraction
How do we know the atomic scale structure
of matter around us?

A crystal is a very large number of atoms or molecules arranged in a periodic

fashion. It acts like a grating with an extremely large number (~Avagadro’s
number) of units that diffract waves coherently.

Every crystal has its own “signature” of the various spacings between atoms.

By measuring the diffraction, we can determine the atomic scale structure.

Typical distances between atoms are of order 0.1-0.3 nm (1-3×10-10 m).

What characteristic wavelengths are needed to study crystals?

We want:
• λ < spacing (so that we can get δ > λ).
• λ not too small (so that θ isn’t too small).
That is: λ~10-10m. ⇒ X-rays!

Lecture 6, p. 14
 FYI: Crystal diffraction
The structure of the crystal can be found using almost the same law
we have for optical gratings!
Bragg Law for constructive interference: δ = 2d sinθ = mλ
d = lattice spacing, λ = x-ray wavelength
θ = x-ray angle (with respect to plane of crystal) Example of
planes in a
NaCl crystal
d

Each crystal has many values of d - the distances between different planes. For a
known wavelength λ the observed angles θ can be used to determine the crystal
structure.
Lecture 6, p. 15
 FYI: Application: Structure of DNA
How do we know the structures of DNA, proteins and other biological
molecules? X-ray diffraction!
The molecules are crystallized to create a crystal in which the molecules are
arranged in a periodic lattice. By using the sharp Bragg diffraction from many
molecules, the structure of each molecule is determined - the positions of
thousands of atoms
The original diffraction image of DNA
taken by Rosalind Franklin in 1953
Actually the DNA was not crystallized for
the first DNA images, but the DNA was
dehydrated in a fiber, and formed a
“quasi-crystalline structure that showed
the helical structure.
The dark bands arranged in a cross
were the first evidence of the helical
structure.

See figure and more details on this and

other X-ray scattering in the text, sec. 36.6
Lecture 6, p. 16
 FYI: Modern Applications in Biology

X-rays remain the primary methods for establishing the atomic

scale structures of complex molecules.

Example: Rabbit liver carboxylesterase

(one molecule showing atomic groups and
attached large scale structures with atoms
not shown).
“Alternative strategies to improve the
antitumor efficacy have concentrated
upon the design of novel camptothecin
analogs. To effect this, we have
determined the x-ray crystal structures
of the rabbit liver carboxylesterase …”

Source: St. Jude Children Research Hospital

http://www.stjude.org/faculty/0,2512,407_2030_4278,00.html
Lecture 6, p. 17
 Diffraction from gratings

Resolving power of an N-slit grating: The Rayleigh criterion

The slit/line spacing determines the location of the peaks (and the
angular dispersing power θ(λ) of the grating.
The number of slits/beam size determines the width of the peaks
(narrower peaks easier to resolve).

Lecture 6, p. 18
 Diffraction Gratings (1)
Diffraction gratings rely on N-slit interference.
They consist of a large number of evenly spaced parallel slits.
An important question:
How effective are diffraction gratings at resolving light of different
wavelengths (i.e. separating closely-spaced ‘spectral lines’)?

IN = N2I1

λ1 sinθ depends on λ.
0 λ1/d sin θ

λ2
0 λ2/d sin θ

Example: Na has a spectrum with two yellow “lines” very

close together: λ1 = 589.0 nm, λ2 = 589.6 nm (∆λ = 0.6 nm)
Are these two lines distinguishable using a particular grating?
We need a “resolution criterion”. Lecture 6, p. 19
 Diffraction Gratings (2)
We use Rayleigh’s criterion:
The minimum wavelength separation we can resolve occurs
when the λ2 peak coincides with the first zero of the λ1 peak:
λ1/d λ2/d
So, the Raleigh criterion is ∆(sinθ)min = λ/Nd.
However, the location of the peak is sinθ = mλ/d.
Thus, (∆λ)min = (d/m)∆(sinθ)min = λ/mN: λ
∆λmin 1 Nd sinθ
=
λ Nm
Comments:
• It pays to use a grating that has a large number of lines, N.
However, one must illuminate them all to get this benefit.
• It also pays to work at higher order (larger m): The widths of
the peaks don’t depend on m, but they are farther apart at large m.
First order Second order Third order
m=1 m=2 m=3

0 λ/d 2λ/d 3λ/d sin θ

Lecture 4, p 20
 ACT 2
1. Suppose we fully illuminate a grating for which d = 2.5 µm.
How big must it be to resolve the Na lines (589 nm, 589.6 nm),
if we are operating at second order (m = 2)?
a. 0.12 mm b. 1.2 mm c. 12 mm

2. How many interference orders can be seen with this grating?

a. 2 b. 3 c. 4

3. Which will reduce the maximum number of interference orders?

a. Increase λ b. Increase d c. Increase N

Lecture 6, p. 21
 Solution
1. Suppose we fully illuminate a grating for which d = 2.5 µm.
How big must it be to resolve the Na lines (589 nm, 589.6 nm),
if we are operating at second order (m = 2)?
a. 0.12 mm b. 1.2 mm c. 12 mm

We need to make N large enough to satisfy Raleigh’s criterion.

∆λ 1 λ 589nm
= So: N ≥ = = 491
λ Nm m ∆λ 2(0.6nm)
Size = Nd ≥ 491×2.5 µm ≈ 1.2 mm

2. How many interference orders can be seen with this grating?

a. 2 b. 3 c. 4

Lecture 6, p. 22
 Solution
1. Suppose we fully illuminate a grating for which d = 2.5 µm.
How big must it be to resolve the Na lines (589 nm, 589.6 nm),
if we are operating at second order (m = 2)?
a. 0.12 mm b. 1.2 mm c. 12 mm

2. How many interference orders can be seen with this grating?

a. 2 b. 3 c. 4

The sin of the diffraction angle can never be larger than 1: sinθ ≤ 1.
From sinθ = mλ/d, we obtain m ≤ d/λ = 2.5 µm/0.589 µm = 4.2.
So, m = 4.

3. Which will reduce the maximum number of interference orders?

a. Increase λ b. Increase d c. Increase N

Lecture 6, p. 23
 Solution
1. Assuming we fully illuminate the grating from the previous
problem (d = 2.5 µm), how big must it be to resolve the Na lines
(589 nm, 589.6 nm)?
a. 0.12 mm b. 1.2 mm c. 12 mm

2. How many interference orders can be seen with this grating?

a. 2 b. 3 c. 4

3. Which will reduce the maximum number of interference orders?

a. Increase λ b. Increase d c. Increase N

m ≤ d/λ, so increase λ, or decrease d.

Changing N does not affect the number of orders.

Lecture 6, p. 24
 Angular Resolution (from L5)
Diffraction also limits our ability to “resolve” (i.e., distinguish) two point sources.
Consider two point sources (e.g., stars) with angular separation α viewed
through a circular aperture or lens of diameter D.
Two point
sources D Just as before,
α
Rayleigh’s Criterion
α
define the images to be
resolved if the central
maximum of one image
α=2αc α=αc α=αc/3 falls on the first minimum
I0 I0 I0 of the second image.

I0.5 I I
λ
α c = 1.22
0 0 0 D
2I0 2I0 2I0

Sum Sum Sum

0 0 0
y y y NOTE:
Two images No interference!!
‘Diffraction limit’ Two images
Why not?
resolvable of resolution not resolvable
Lecture 6, p. 25
 Exercise: Angular resolution
Car headlights in the distance:
What is the maximum distance L you can
be from an oncoming car at night, and still
distinguish its two headlights, which are
separated by a distance d = 1.5 m?
Assume that your pupils have a diameter
D = 2 mm at night, and that the wavelength
of light is λ = 550 nm.

Lecture 6, p. 26
 Solution
Car headlights in the distance:
What is the maximum distance L you can
be from an oncoming car at night, and still
distinguish its two headlights, which are
separated by a distance d = 1.5 m?
Assume that your pupils have a diameter
D = 2 mm at night, and that the wavelength
of light is λ = 550 nm.

λ
Use Rayleigh’s criterion: α c = 1.22 = 3.4 × 10 −4 (radians)
D

Then, L ≈ d/αc = 4500 m = 2.8 miles (assuming perfect eyes).

The small angle approximation is valid.

Lecture 6, p. 27
 Michelson Interferometer
(from Lect. 5)
The Michelson interferometer works by varying the phase difference
between the two paths the light can take.

One possibility is to vary the lengths L1 or L2.

This makes possible very accurate measurements of displacements.

mirror Path-length Be careful !!

Total path difference What’s important
length L1 δ = L2 - L1 is the total time to
get back to the
2I1 2I1 beam splitter...
Total path
Source length L2

4I1 2I1
mirror
2I1
beam I1 I1
splitter

I = 4I1 cos2(φ/2), with φ = 2π δ/λ

Lecture 6, p. 28
 Exercise: Michelson Interferometer
An interferometer is sensitive to the phase
difference, φ, between the two paths. One way
to vary φ is to change L1 or L2. Another way is mirrors
L1
to change the index of refraction of the medium Source
in one arm.
beam
Suppose that in our interferometer, the mirrors splitter L2
are 1 m from the beam splitter (L1 = L2 = 2 m).
Remember: L1 and L2 are
Also, assume the index of refraction in arm 1 is detector the round trip distances.
n1 = 1.0. What is the smallest deviation of n2
from 1.0 that will give destructive interference
when the beams are recombined? Assume
that λ = 600 nm in vacuum.

Lecture 6, p. 29
 Solution
An interferometer is sensitive to the phase
difference, φ, between the two paths. One way
to vary φ is to change L1 or L2. Another way is mirrors
L1
to change the index of refraction of the medium Source
in one arm.
beam
Suppose that in our interferometer, the mirrors splitter L2
are 1 m from the beam splitter (L1 = L2 = 2 m). Box of gas with n2.
Also, assume the index of refraction in arm 1 is detector
n1 = 1.0. What is the smallest deviation of n2
from 1.0 that will give destructive interference
when the beams are recombined? Assume
that λ = 600 nm in vacuum.
We want the number of wavelengths in arm 2 to differ from that in arm 1 by ½.
The number in arm 1 is N1 = L1/λ = 3333333.3, so we want N2 = 3333333.8.
The ratio, N2/N1 = 1+1.5×10-7 is the ratio of wavelengths, which, by definition, is
the index of refraction in the gas. This instrument is very sensitive to small
changes in n. For comparison nair ~ 1+3×10-4.

Lecture 6, p. 30