?

Flies have compound eyes with

­thousands of miniature lenses. The
­overall diameter of the eye is about 1 mm,
but each lens is only about 20 μm in
­diameter and produces an individual image
of a small region in the fly’s field of view.
Compared to the resolving power of the
human eye (in which the light-gathering
­region is about 16 mm across), the ability
of a fly’s eye to resolve small details is
(i) worse because the lenses are so small;
(ii) worse because the eye as a whole is so
small; (iii) better because the lenses are
so small; (iv) better because the eye as a
whole is so small; (v) about the same.

36 Diffraction
E
veryone is used to the idea that sound bends around corners. If sound didn’t behave LEARNING OUTCOMES
this way, you couldn’t hear a police siren that’s out of sight or the speech of a person In this chapter, you’ll learn…
whose back is turned to you. But light can bend around corners as well. When light 36.1 What happens when coherent light
from a point source falls on a straightedge and casts a shadow, the edge of the shadow is shines on an object with an edge or
never perfectly sharp. Some light appears in the area that we expect to be in the shadow, aperture.
and we find alternating bright and dark fringes in the illuminated area. In general, light 36.2 How to understand the diffraction
emerging from apertures doesn’t precisely follow the predictions of the straight-line ray ­pattern formed when coherent light
passes through a narrow slit.
model of geometric optics. 36.3 How to calculate the intensity at various
The reason for these effects is that light, like sound, has wave characteristics. In points in a single-slit diffraction pattern.
Chapter 35 we studied the interference patterns that can arise when two light waves are 36.4 What happens when coherent light
combined. In this chapter we’ll investigate interference effects due to combining many shines on an array of narrow, closely
light waves. Such effects are referred to as diffraction. The behavior of waves after they spaced slits.
36.5 How scientists use diffraction
pass through an aperture is an example of diffraction; each infinitesimal part of the aper- ­gratings for precise measurements of
ture acts as a source of waves, and these waves interfere, producing a pattern of bright and wavelength.
dark fringes. 36.6 How x-ray diffraction reveals the
Similar patterns appear when light emerges from arrays of apertures. The nature of ­arrangement of atoms in a crystal.
36.7 How diffraction sets limits on the
these patterns depends on the color of the light and the size and spacing of the apertures.
­smallest details that can be seen with
Examples of this effect include the colors of iridescent butterflies and the “rainbow” an optical system.
you see reflected from the surface of a compact disc. We’ll explore similar effects with 36.8 How holograms work.
x rays that are used to study the atomic structure of solids and liquids. Finally, we’ll look
at the physics of a hologram, a special kind of interference pattern used to form three-­ You’ll need to review…
33.4, 33.7 Prisms and dispersion; Huygens’s
dimensional images.
principle.
34.4, 34.5 Image formation by a lens;
f-number.
36.1 FRESNEL AND FRAUNHOFER DIFFRACTION 35.1–35.3 Coherent light, two-slit interference,
and phasors.
According to geometric optics, when an opaque object is placed between a point light
source and a screen, as in (Fig. 36.1, next page) the shadow of the object forms a perfectly
sharp line. No light at all strikes the screen at points within the shadow, and the area outside
the shadow is illuminated nearly uniformly. But as we saw in Chapter 35, the wave nature
of light causes effects that can’t be understood with geometric optics. An important class of
such effects occurs when light strikes a barrier that has an aperture or an edge. The interfer-
ence patterns formed in such a situation are grouped under the heading diffraction.
1185
1186    CHAPTER 36 Diffraction

Figure 36.1 A point source of light Figure 36.2 shows an example of diffraction. The photograph in Fig. 36.2a was made
illuminates a straightedge. by placing a razor blade halfway between a pinhole, illuminated by monochromatic light,
Geometric optics predicts that this situation and a photographic film. The film recorded the shadow cast by the blade. Figure 36.2b is
should produce a sharp boundary between an enlargement of a region near the shadow of the right edge of the blade. The position of
illumination and
solid shadow.
the geometric shadow line is indicated by arrows. The area outside the geometric shadow
DOESN’T is bordered by alternating bright and dark bands. There is some light in the shadow r­ egion,
That’s NOT what HAPPEN although this is not very visible in the photograph. The first bright band in Fig. 36.2b, just
really happens! to the right of the geometric shadow, is considerably brighter than in the region of uniform
Point Area of
illumination to the extreme right. This simple experiment gives us some idea of the rich-
source illumination ness and complexity of diffraction.
We don’t often observe diffraction patterns such as Fig. 36.2 in everyday life because
most ordinary light sources are neither monochromatic nor point sources. If we use a
Geometric white frosted light bulb instead of a point source to illuminate the razor blade in Fig. 36.2,
shadow each wavelength of the light from every point of the bulb forms its own diffraction pat-
Straightedge tern, but the patterns overlap so much that we can’t see any individual pattern.
Screen
Diffraction and Huygens’s Principle
We can use Huygens’s principle (see Section 33.7) to analyze diffraction patterns. This
­principle states that we can consider every point of a wave front as a source of secondary
wavelets. These spread out in all directions with a speed equal to the speed of propagation
of the wave. The position of the wave front at any later time is the envelope of the secondary
wavelets at that time. To find the resultant displacement at any point, we use the superposition
principle to combine all the individual displacements produced by these secondary waves.
In Fig. 36.1, both the point source and the screen are relatively close to the obstacle
forming the diffraction pattern. This situation is described as near-field diffraction or
Fresnel diffraction, pronounced “Freh-nell” (after the French scientist Augustin Jean
Fresnel, 1788–1827). By contrast, we use the term Fraunhofer diffraction (after the
German physicist Joseph von Fraunhofer, 1787–1826) for situations in which the source,
obstacle, and screen are far enough apart that we can consider all lines from the source to
the obstacle to be parallel, and can likewise consider all lines from the obstacle to a given
point on the screen to be parallel. We’ll restrict the following discussion to Fraunhofer dif-
fraction, which is usually simpler to analyze in detail than Fresnel diffraction.
Diffraction is sometimes described as “the bending of light around an obstacle.” But the
process that causes diffraction is present in the propagation of every wave. When part of
the wave is cut off by some obstacle, we observe diffraction effects that result from inter-
ference of the remaining parts of the wave fronts. Optical instruments typically use only
a limited portion of a wave; for example, a telescope uses only the part of a wave that is
admitted by its objective lens or mirror. Thus diffraction plays a role in nearly all optical
phenomena.

Figure 36.2 An example of diffraction.

(a) (b)
Photograph of a razor blade illuminated by monochromatic
light from a point source (a pinhole). Notice the fringe
around the blade outline.

Enlarged view of the area outside

the blade’s geometric shadow.

Position of geometric shadow

 36.2 Diffraction from a Single Slit     1187

Finally, we emphasize that there is no fundamental distinction between interference

and diffraction. In Chapter 35 we used the term interference for effects involving waves
from a small number of sources, usually two. Diffraction usually involves a continuous
distribution of Huygens’s wavelets across the area of an aperture, or a very large number
of sources or apertures. But both interference and diffraction are consequences of super-
position and Huygens’s principle.

TEST YOUR UNDERSTANDING OF SECTION 36.1 Can sound waves undergo diffraction
around an edge?

ANSWER
sounds only from objects that were in plain view.
that underwent diffraction. If there were no diffraction or reflection of sound, you could hear
❙ yes When you hear the voice of someone standing around a corner, you are hearing sound waves

36.2 DIFFRACTION FROM A SINGLE SLIT

In this section we’ll discuss the diffraction pattern formed by plane-wave (parallel-ray)
monochromatic light when it emerges from a long, narrow slit, as shown in Fig. 36.3. We
call the narrow dimension the width, even though in this figure it is a vertical dimension.
According to geometric optics, the transmitted beam should have the same cross
section as the slit, as in Fig. 36.3a. What is actually observed is the pattern shown in
Fig. 36.3b. The beam spreads out vertically after passing through the slit. The diffraction
pattern consists of a central bright band, which may be much broader than the width of the
slit, bordered by alternating dark and bright bands with rapidly decreasing intensity. About
85% of the power in the transmitted beam is in the central bright band, whose width is
inversely proportional to the slit width. In general, the narrower the slit, the broader the
entire diffraction pattern. (The horizontal spreading of the beam in Fig. 36.3b is negli-
gible because the horizontal dimension of the slit is relatively large.) You can observe
a similar diffraction pattern by looking at a point source, such as a distant street light,
through a narrow slit formed between your two thumbs held in front of your eye; the retina
of your eye acts as the screen.

Single-Slit Diffraction: Locating the Dark Fringes

Figure 36.4 (next page) shows a side view of the same setup; the long sides of the slit are
perpendicular to the figure, and plane waves are incident on the slit from the left. According
to Huygens’s principle, each element of area of the slit opening can be considered as a source
of secondary waves. In particular, imagine dividing the slit into several narrow strips of equal
width, parallel to the long edges and perpendicular to the page. Figure 36.4a shows two such
strips. Cylindrical secondary wavelets, shown in cross section, spread out from each strip.

Figure 36.3 (a) The “shadow” of a horizontal slit as incorrectly predicted by geometric optics.
(b) A horizontal slit actually produces a diffraction pattern. The slit width has been greatly exaggerated.
(a) PREDICTED OUTCOME: (b) WHAT REALLY HAPPENS:
Geometric optics predicts that this In reality, we see a diffraction
setup will produce a single bright Screen pattern—a set of bright and
band the same size as the slit. dark fringes.

a a

Width

Parallel-ray monochromatic
light
1188    CHAPTER 36 Diffraction

Figure 36.4 Diffraction by a single rectangular slit. The long sides of the slit are perpendicular to the figure.
(a) A slit as a source of wavelets (b) Fresnel (near-field) diffraction (c) Fraunhofer (far-field) diffraction (d) Imaging Fraunhofer diffraction

We divide the slit into If the screen is close, If the screen is distant, A converging lens images a
imaginary strips parallel the rays from the the rays to P are Fraunhofer pattern on
to the slit’s long axis. different strips to a approximately parallel. a nearby screen.
point P on the screen
Slit are not parallel. Converging
width cylindrical lens
a a

Each strip is a source of P P

Huygens’s wavelets.
Plane waves f
incident on the slit Screen Screen

In Fig. 36.4b a screen is placed to the right of the slit. We can calculate the resultant
intensity at a point P on the screen by adding the contributions from the individual wave-
lets, taking proper account of their various phases and amplitudes. It’s easiest to do this
calculation if we assume that the screen is far enough away that all the rays from vari-
ous parts of the slit to a particular point P on the screen are parallel, as in Fig. 36.4c. An
equivalent situation is Fig. 36.4d, in which the rays to the lens are parallel and the lens
forms a reduced image of the same pattern that would be formed on an infinitely distant
screen without the lens. We might expect that the various light paths through the lens
would introduce additional phase shifts, but in fact it can be shown that all the paths have
equal phase shifts, so this is not a problem.
The situation of Fig. 36.4b is Fresnel diffraction; those in Figs. 36.4c and 36.4d, where
the outgoing rays are considered parallel, are Fraunhofer diffraction. We can derive quite
simply the most important characteristics of the Fraunhofer diffraction pattern from a
single slit. First consider two narrow strips, one just below the top edge of the drawing of
the slit and one at its center, shown in end view in Fig. 36.5. The difference in path length
to point P is 1a>22 sin u, where a is the slit width and u is the angle between the perpen-
dicular to the slit and a line from the center of the slit to P. Suppose this path difference
happens to be equal to l>2; then light from these two strips arrives at point P with a half-
cycle phase difference, and cancellation occurs.
Similarly, light from two strips immediately below the two in the figure also arrives
at P a half-cycle out of phase. In fact, the light from every strip in the top half of the slit
cancels out the light from a corresponding strip in the bottom half. Hence the combined

Figure 36.5 Side view of a horizontal slit. (a)

When the distance x to the screen is much P
greater than the slit width a, the rays from y
a distance a>2 apart may be considered a x
u
parallel. O
For the two strips shown, the path difference to P is 1a>22 sin u.
When 1a>22 sin u = l>2, the light cancels at P. This is true for the
whole slit, so P represents a dark fringe.

(b) Enlarged view of the top half of the slit

u is usually very small, so we can use the

u approximations sin u = u and tan u = u.
a
u Then the condition for a dark band is
2
ml
ym = x
a a
sin u
2
 36.2 Diffraction from a Single Slit     1189

light from the entire slit completely cancels at P, giving a dark fringe in the interference
pattern. A dark fringe occurs whenever
a l l
sin u = { or sin u = { (36.1)
2 2 a
The plus-or-minus 1{2 sign in Eq. (36.1) says that there are symmetric dark fringes above and
below point O in Fig. 36.5a. The upper fringe 1u 7 02 occurs at a point P where light from the
bottom half of the slit travels l>2 farther to P than does light from the top half; the lower fringe
1u 6 02 occurs where light from the top half travels l>2 farther than light from the bottom half.
We may also divide the slit into quarters, sixths, and so on, and use the above argument
to show that a dark fringe occurs whenever sin u = {2l>a, {3l>a, and so on. Thus the
condition for a dark fringe is

Dark fringes, Angle of line from center of slit to mth dark fringe on screen
single-slit ml
diffraction: sin u = 1m = {1, {2, {3, c2 (36.2)
a
Slit width Wavelength

For example, if the slit width is equal to ten wavelengths 1a = 10l2, dark fringes occur
at sin u = { 101
, { 10
2
, { 10
3
, c. Between the dark fringes are bright fringes. Note that
sin u = 0 corresponds to a bright band; in this case, light from the entire slit arrives at P
in phase. Thus it would be wrong to put m = 0 in Eq. (36.2).
With light, the wavelength l is of the order of 500 nm = 5 * 10-7 m. This is often
much smaller than the slit width a; a typical slit width is 10-2 cm = 10-4 m. Therefore
the values of u in Eq. (36.2) are often so small that the approximation sin u ≈ u (where u
is in radians) is a very good one. In that case we can rewrite this equation as
ml
u = (m = {1, {2, {3, c) (for small angles u in radians) Figure 36.6 Photograph of the
a Fraunhofer diffraction pattern of a
­single horizontal slit.
Also, if the distance from slit to screen is x, as in Fig. 36.5a, and the vertical distance of
the mth dark band from the center of the pattern is ym , then tan u = ym>x. For small u we
m = 3
may also approximate tan u by u (in radians). We then find
m = 2
ml
ym = x 1for ym V x2(36.3)
a m = 1
Figure 36.6 is a photograph of a single-slit diffraction pattern with the m = {1, {2,
and {3 minima labeled. The central bright fringe is wider than the other bright fringes;
in the small-angle approximation used in Eq. (36.3), it is exactly twice as wide.
m = -1
   CAUTION    Single-slit diffraction vs. two-slit interference Equation (36.3) has the same form as
the equation for the two-slit pattern, Eq. (35.6), except that in Eq. (36.3) we use x rather than R for m = -2
the distance to the screen. But Eq. (36.3) gives the positions of the dark fringes in a single-slit pat-
m = -3
tern rather than the bright fringes in a double-slit pattern. Also, m = 0 in Eq. (36.2) is not a dark
fringe. Finally, note that Eq. (36.2) gives the angle u from the mth dark fringe to the center of the
diffraction pattern—not the angle from a dark fringe on one side of the pattern to the corresponding
dark fringe on the other side. Be careful! ❙

EXAMPLE 36.1 Single-slit diffraction WITH ARIATION PROBLEMS

You pass 633 nm laser light through a narrow slit and observe the dif- EXECUTE The first minimum corresponds to m = 1 in Eq. (36.3).
fraction pattern on a screen 6.0 m away. The distance on the screen be- The distance y1 from the central maximum to the first minimum
tween the centers of the first minima on either side of the central bright on either side is half the distance between the two first minima, so
fringe is 32 mm (Fig. 36.7, next page). How wide is the slit? y1 = 132 mm2>2 = 16 mm. Solving Eq. (36.3) for a, we find
IDENTIFY and SET UP This problem involves the relationship between xl 16.0 m21633 * 10-9 m2
the positions of dark fringes in a single-slit diffraction pattern and the slit a = = = 2.4 * 10-4 m = 0.24 mm
y1 16 * 10-3 m
width a (our target variable). The distances between fringes on the screen
are much smaller than the slit-to-screen distance, so the angle u shown in
Fig. 36.5a is very small and we can use Eq. (36.3) to solve for a. Continued
1190    CHAPTER 36 Diffraction

Figure 36.7 A single-slit diffraction experiment. EVALUATE The angle u is small only if the wavelength is small
y c­ ompared to the slit width. Since l = 633 nm = 6.33 * 10-7 m and
we have found a = 0.24 mm = 2.4 * 10-4 m, our result is consis-
tent with this: The wavelength is 16.33 * 10-7 m2 > (2.4 * 10-4 m) =
Slit width = ? 0.0026 as large as the slit width. Can you show that the distance
­between the second minima on either side is 2132 mm2 = 64 mm,
and so on?
32 mm x
KEYCONCEPT In single-slit diffraction, wavelets that emerge from
different parts of the slit interfere and produce a pattern of bright and
dark fringes. Equations (36.2) and (36.3) give the positions of the dark
x = 6.0 m fringes where destructive interference occurs.
Screen

Figure 36.8 Using phasorS diagrams to TEST YOUR UNDERSTANDING OF SECTION 36.2 Rank the following single-slit diffraction
find the amplitude of the E field in single- experiments in order of the size of the angle from the center of the diffraction pattern to the first
slit
S
diffraction. Each phasor represents the dark fringe, from largest to smallest: (i) Wavelength 400 nm, slit width 0.20 mm; (ii) wavelength
E field from a single strip within the slit. 600 nm, slit width 0.20 mm; (iii) wavelength 400 nm, slit width 0.30 mm; (iv) wavelength 600 nm,
(a) slit width 0.30 mm.

(iv) 1600 nm2>10.30 mm2 = 16.0 * 10-7 m2>13.0 * 10-4 m2 = 2.0 * 10-3.
Strips within slit

(iii) 1400 nm2>10.30 mm2 = 14.0 * 10-7 m2>13.0 * 10-4 m2 = 1.3 * 10-3;
Slit ANSWER
(ii) 1600 nm2>10.20 mm2 = 16.0 * 10-7 m2>12.0 * 10-4 m2 = 3.0 * 10-3;
width O

(i) 1400 nm2>10.20 mm2 = 14.0 * 10-7 m2>12.0 * 10-4 m2 = 2.0 * 10-3;
a
P
Distant screen the value of u. The ratio l>a in each case is
Plane waves m = 1, or sin u = l>a. The larger the value of the ratio l>a, the larger the value of sin u and hence
incident on the slit ❙ (ii), (i) and (iv) (tie), (iii) The angle u of the first dark fringe is given by Eq. (36.2) with

(b) At the center of the diffraction pattern

(point O), the phasors from all strips within the
slit are in phase.
36.3 INTENSITY IN THE SINGLE-SLIT PATTERN
E0
We can derive an expression for the intensity distribution for the single-slit diffraction
pattern by the same phasor-addition method that we used in Section 35.3 for the two-
(c) Phasor diagram at a point slightly off the slit interference pattern. We again imagine a plane wave front at the slit subdivided into
center of the pattern; b = total phase difference a large number of strips. We superpose the contributions of the Huygens wavelets from
between the first and last phasors. all the strips at a point P on a distant screen at an angle u from the normal to the slit S
plane (Fig. 36.8a). To do this, we use a phasor to represent the sinusoidally varying E
field from each strip. The magnitude
S
of the vector sum of the phasors at each point P
EP is the amplitude EP of the total E field at that point. The intensity at P is proportional
to EP 2.
At the point O shown in Fig. 36.8a, corresponding to the center of the pattern where
E0 b
u = 0, there are negligible path differences for x W a; the phasors are all essentially in
phase (that is, have the same direction). In Fig. 36.8b we draw the phasors at time t = 0
and denote the resultant amplitude at O by E0 . In this illustration we have divided the slit
(d) As in (c), but in the limit that the slit is
into 14 strips.
subdivided into infinitely many strips
Now consider wavelets arriving from different strips at point P in Fig. 36.8a, at an
C
angle u from point O. Because of the differences in path length, there are now phase
E0 differences between wavelets coming from adjacent strips; the corresponding phasor dia-
b b gram is shown in Fig. 36.8c. The vector sum of the phasors is now part of the perimeter
b
2 of a many-sided polygon, and EP , the amplitude of the resultant electric field at P, is the
E0 b
b chord. The angle b is the total phase difference between the wave received at P from the
b 2 B
E 0 si
n 2 top strip of Fig. 36.8a and the wave received at P from the bottom strip.
b
b We may imagine dividing the slit into narrower and narrower strips. In the limit that
E 0 si
n 2 EP there is an infinite number of infinitesimally narrow strips, the curved trail of phasors be-
b comes an arc of a circle (Fig. 36.8d), with arc length equal to the length E0 in Fig. 36.8b.
E0 b
A The center C of this arc is found by constructing perpendiculars at A and B. From the re-
lationship among arc length, radius, and angle, the radius of the arc is E0>b; the amplitude
 36.3 Intensity in the Single-Slit Pattern     1191

EP of the resultant electric field at P is equal to the chord AB, which is 21E0>b2 sin1b>22.
(Note that b must be in radians!) We then have

sin1b>22
EP = E0 (amplitude in single@slit diffraction) (36.4)
b>2
The intensity at each point on the screen is proportional to the square of the amplitude
given by Eq. (36.4). If I0 is the intensity in the straight-ahead direction where u = 0 and
b = 0, then the intensity I at any point is

sin1b>22 2
I = I0 c d (intensity in single@slit diffraction) (36.5)
b>2
We can express the phase difference b in terms of geometric quantities, as we did for
the two-slit pattern. From Eq. (35.11) the phase difference is 2p>l times the path differ-
ence. Figure 36.5 shows that the path difference between the ray from the top of the slit
and the ray from the middle of the slit is 1a>22 sin u. The path difference between the rays
from the top of the slit and the bottom of the slit is twice this, so
2p
b = a sin u (36.6)
l
and Eq. (36.5) becomes

Angle of line from center of slit to position on screen

Intensity in sin 3pa1sin u2>l4 2 Figure 36.9 (a) Intensity versus angle
single-slit I = I0 e f (36.7) in single-slit diffraction. The values of m
diffraction pa1sin u2>l label intensity minima given by Eq. (36.8).
Intensity at u = 0 Slit width Wavelength Most of the wave power goes into the
central intensity peak (between the m = 1
and m = - 1 intensity minima). (b) These
This equation expresses the intensity directly in terms of the angle u. In many calculations water waves passing through a small ap-
it is easier first to calculate the phase angle b, from Eq. (36.6), and then to use Eq. (36.5). erture behave exactly like light waves in
single-slit diffraction. Only the diffracted
Equation (36.7) is plotted in Fig. 36.9a. Note that the central intensity peak is much waves within the central intensity peak are
larger than any of the others. This means that most of the power in the wave remains visible; the waves at larger angles are too
within an angle u from the perpendicular to the slit, where sin u = l>a (the first diffrac- faint to see.
tion minimum). You can see this easily in Fig. 36.9b, which is a photograph of water (a) u
waves undergoing single-slit diffraction. Note also that the peak intensities in Fig. 36.9a I = 0.0083I0
m = 3
decrease rapidly as we go away from the center of the pattern. (Compare Fig. 36.6, which
I = 0.0165I0
shows a single-slit diffraction pattern for light.) m = 2
The dark fringes in the pattern are the places where I = 0. These occur at points for I = 0.0472I0
which the numerator of Eq. (36.5) is zero so that b is a multiple of 2p. From Eq. (36.6) m = 1
this corresponds to u
O
a sin u I = I0
= m 1m = {1, {2, c2 m = -1
l
ml m = -2
sin u = 1m = {1, {2, c2(36.8)
a m = -3
This agrees with our previous result, Eq. (36.2). Note again that b = 0 (corresponding to
u = 0) is not a minimum. Equation (36.5) is indeterminate at b = 0, but we can evaluate the
(b)
limit as b S 0 by using L’Hôpital’s rule. We find that at b = 0, I = I0 , as we should expect.

Intensity Maxima in the Single-Slit Pattern

We can also use Eq. (36.5) to calculate the positions of the peaks, or intensity maxima,
and the intensities at these peaks. This is not quite as simple as it may appear. We might
expect the peaks to occur where the sine function reaches the value {1—namely, where
b = {p, {3p, {5p, or in general,

b ≈ {12m + 12p 1m = 0, 1, 2, c2 (36.9)

1192    CHAPTER 36 Diffraction

This is approximately correct, but because of the factor 1b>22 2 in the denominator of
Eq. (36.5), the maxima don’t occur precisely at these points. When we take the derivative
of Eq. (36.5) with respect to b and set it equal to zero to try to find the maxima and min-
ima, we get a transcendental equation that has to be solved numerically. In fact there is no
maximum near b = {p. The first maxima on either side of the central maximum, near
b = {3p, actually occur at {2.860 p. The second side maxima, near b = {5p, are
actually at {4.918p, and so on. The error in Eq. (36.9) vanishes in the limit of large m—
that is, for intensity maxima far from the center of the pattern.
To find the intensities at the side maxima, we substitute these values of b back into
Eq. (36.5). Using the approximate expression in Eq. (36.9), we get

I0
Im ≈ (36.10)
1m + 1
2 2 2p 2
where Im is the intensity of the mth side maximum and I0 is the intensity of the central
maximum. Equation (36.10) gives the series of intensities

0.0450I0 0.0162I0 0.0083I0

and so on. As we have pointed out, this equation is only approximately correct. The actual
intensities of the side maxima turn out to be

0.0472I0 0.0165I0 0.0083I0 c

These intensities decrease very rapidly, as Fig. 36.9a also shows. Even the first side
­maxima have less than 5% of the intensity of the central maximum.

Width of the Single-Slit Pattern

For small angles the angular spread of the diffraction pattern is inversely proportional to
the ratio of the slit width a to the wavelength l. Figure 36.10 shows graphs of intensity I
as a function of the angle u for three values of the ratio a>l.
With light waves, the wavelength l is often much smaller than the slit width a, and the
values of u in Eqs. (36.6) and (36.7) are so small that the approximation sin u = u is very
good. With this approximation the position u1 of the first 1m = 12 minimum, correspond-
ing to b>2 = p, is, from Eq. (36.7),
l
u1 = (36.11)
a

This characterizes the width (angular spread) of the central maximum, and we see that it
is inversely proportional to the slit width a. When the small-angle approximation is valid,
the central maximum is exactly twice as wide as each side maximum. When a is of the
order of a centimeter or more, u1 is so small that we can consider practically all the light to
be concentrated at the geometrical focus. But when a is less than l, the central maximum
spreads over 180°, and the fringe pattern is not seen at all.

Figure 36.10 The single-slit diffraction pattern depends on the ratio of the slit width a to the wavelength l.
(a) a = l (b) a = 5l (c) a = 8l

If the slit width is equal to or narrower than the The wider the slit (or the shorter the
wavelength, only one broad maximum forms. wavelength), the narrower and sharper
I I is the central peak. I
I0 I0 I0

u u u
-20° -10° 0° 10° 20° -20° -10° 0° 10° 20° -20° -10° 0° 10° 20°
 36.3 Intensity in the Single-Slit Pattern     1193

It’s important to keep in mind that diffraction occurs for all kinds of waves, not just Figure 36.11 The sound waves used in
light. Sound waves undergo diffraction when they pass through a slit or aperture such as speech have a long wavelength (about 1 m)
and can easily bend around this instructor’s
an ordinary doorway. The sound waves used in speech have wavelengths of about a meter
head. By contrast, light waves have very short
or greater, and a typical doorway is less than 1 m wide; in this situation, a is less than l, wavelengths and undergo very little diffrac-
and the central intensity maximum extends over 180°. This is why the sounds coming tion. Hence you can’t see around his head!
through an open doorway can easily be heard by an eavesdropper hiding out of sight
around the corner. In the same way, sound waves can bend around the head of an instruc-
tor who faces the blackboard while lecturing (Fig. 36.11). By contrast, there is essentially
no diffraction of visible light through a doorway because the width a is very much greater
than the wavelength l (of order 5 * 10-7 m). You can hear around corners because typi-
cal sound waves have relatively long wavelengths; you cannot see around corners because
the wavelength of visible light is very short.

EXAMPLE 36.2 Single-slit diffraction: Intensity I WITH ARIATION PROBLEMS

(a) The intensity at the center of a single-slit diffraction pattern is I0 . What is For example, for 550 nm light the slit width is a = 18621550 nm2 =
the intensity at a point in the pattern where there is a 66 radian phase differ- 4.7 * 10-5 m = 0.047 mm, or roughly 20 1
mm.
ence between wavelets from the two edges of the slit? (b) If this point is 7.0°
EVALUATE To what point in the diffraction pattern does this value
away from the central maximum, how many wavelengths wide is the slit?
of b correspond? To find out, note that b = 66 rad is approximately
IDENTIFY and SET UP In our analysis of Fig. 36.8 we used the symbol b for equal to 21p. This is an odd multiple of p, corresponding to the form
the phase difference between wavelets from the two edges of the slit. In part 12m + 12p found in Eq. (36.9) for the intensity maxima. Hence
(a) we use Eq. (36.5) to find the intensity I at the point in the pattern where b = 66 rad corresponds to a point near the tenth 1m = 102 maximum.
b = 66 rad. In part (b) we need to find the slit width a as a multiple of the This is well beyond the range shown in Fig. 36.9a, which shows only
wavelength l so our target variable is a>l. We are given the angular position maxima out to m = {3.
u of the point where b = 66 rad, so we can use Eq. (36.6) to solve for a>l.
KEYCONCEPT The intensity at a given point in a single-slit diffrac-
EXECUTE (a) We have b>2 = 33 rad, so from Eq. (36.5), tion pattern depends on the phase difference b between the wavelets
that arrive at that point from the two edges of the slit [Eqs. (36.6) and
sin133 rad2 2
I = I0 c d = 19.2 * 10-42I0 (36.7)]. The intensity is greatest at the center of the pattern where b = 0;
33 rad the intensity is zero at points where b = {2p, {4p, {6p, . . . .

(b) From Eq. (36.6),

a b 66 rad
= = = 86
l 2p sin u 12p rad2sin 7.0°

EXAMPLE 36.3 Single-slit diffraction: Intensity II WITH ARIATION PROBLEMS

In the experiment described in Example 36.1 (Section 36.2), the inten- EVALUATE Figure 36.9a shows that an intensity this high can occur
sity at the center of the pattern is I0 . What is the intensity at a point on only within the central intensity maximum. This checks out; from
the screen 3.0 mm from the center of the pattern? Example 36.1, the first intensity minimum (m = 1 in Fig. 36.9a) is
132 mm2>2 = 16 mm from the center of the pattern, so the point
IDENTIFY and SET UP This is similar to Example 36.2, except that we
in question here at y = 3 mm does, indeed, lie within the central
are not given the value of the phase difference b at the point in question.
maximum.
We use geometry to determine the angle u for our point and then use
Eq. (36.7) to find the intensity I (the target variable). KEYCONCEPT When light of wavelength l passes through a single
slit of width a, you can write the intensity of the diffracted light at angle
EXECUTE Referring to Fig. 36.5a, we have y = 3.0 mm and x = 6.0 m,
-3 -4 u [Eq. (36.7)] in terms of the ratio 1pa sin u2>l, which is one-half of
so tan u = y>x = 13.0 * 10 m2>16.0 m2 = 5.0 * 10 . This is so
the phase difference b between wavelets coming from the two edges of
small that the values of tan u, sin u, and u (in radians) are all nearly the
the slit. The intensity is greatest at the center of the pattern (u = 0); the
same. Then, from Eq. (36.7),
intensity is zero where 1pa sin u2>l = {p, {2p, {3p, . . . .
pa sin u p12.4 * 10-4 m215.0 * 10-42
= = 0.60
l 6.33 * 10-7 m
sin 0.60 2
I = I0 a b = 0.89I0
0.60
1194    CHAPTER 36 Diffraction

TEST YOUR UNDERSTANDING OF SECTION 36.3 Coherent electromagnetic radiation is

sent through a slit of width 0.0100 mm. For which of the following wavelengths will there be no
points in the diffraction pattern where the intensity is zero? (i) Blue light of wavelength 500 nm;
(ii) infrared light of wavelength 10.6 mm; (iii) microwaves of wavelength 1.00 mm; (iv) ultraviolet
light of wavelength 50.0 nm.

1.00 * 10-3 m) but not for (i) 1l = 500 nm = 5.00 * 10-7 m2 or (iv) 1l = 50.0 nm =
ANSWER
5.00 * 10-8 m2.

so this condition is satisfied for (ii) 1l = 10.6 mm = 1.06 * 10-5 m2 and (iii) (l = 1.00 mm =
pattern at which the intensity is zero (see Fig. 36.10a). The slit width is 0.0100 mm = 1.00 * 10-5 m,
❙ (ii) and (iii) If the slit width a is less than the wavelength l, there are no points in the diffraction

36.4 MULTIPLE SLITS

In Sections 35.2 and 35.3 we analyzed interference from two point sources and from two
very narrow slits; in this analysis we ignored effects due to the finite (that is, nonzero) slit
width. In Sections 36.2 and 36.3 we considered the diffraction effects that occur when
light passes through a single slit of finite width. Additional interesting effects occur when
we have two slits with finite width or when there are several very narrow slits.

Two Slits of Finite Width

Figure 36.12 Finding the intensity
­pattern for two slits of finite width. Let’s take another look at the two-slit pattern in the more realistic case in which the slits
have finite width. If the slits are narrow in comparison to the wavelength, we can assume
(a) Single-slit diffraction pattern for a
slit width a
that light from each slit spreads out uniformly in all directions to the right of the slit. We
I0 I used this assumption in Section 35.3 to calculate the interference pattern described by
Eq. (35.10) or (35.15), consisting of a series of equally spaced, equally intense maxima.
However, when the slits have finite width, the peaks in the two-slit interference pattern
are modulated by the single-slit diffraction pattern characteristic of the width of each slit.
u
md = -2 md = -1 0 md = 1 md = 2 Figure 36.12a shows the intensity in a single-slit diffraction pattern with slit width a.
The diffraction minima are labeled by the integer md = {1, {2, c (“d” for “diffrac-
(b) Two-slit interference pattern for narrow tion”). Figure 36.12b shows the pattern formed by two very narrow slits with distance d
slits whose separation d is four times the between slits, where d is four times as great as the single-slit width a in Fig. 36.12a; that
width of the slit in (a) is, d = 4a. The interference maxima are labeled by the integer mi = 0, {1, {2, c (“i”
I0 for “interference”). We note that the spacing between adjacent minima in the single-slit
pattern is four times as great as in the two-slit pattern. Now suppose we widen each of
the narrow slits to the same width a as that of the single slit in Fig. 36.12a. Figure 36.12c
u shows the pattern from two slits with width a, separated by a distance (between centers)
mi = -8 mi = -4 0 mi = 4 mi = 8 d = 4a. The effect of the finite width of the slits is to superimpose the two patterns—
that is, to multiply the two intensities at each point. The two-slit peaks are in the same
(c) Calculated intensity pattern for two slits positions as before, but their intensities are modulated by the single-slit pattern, which
of width a and separation d = 4a, including
both interference and diffraction effects acts as an “envelope” for the intensity function. The expression for the intensity shown
in Fig. 36.12c is proportional to the product of the two-slit and single-slit expressions,
I0
Eqs. (35.10) and (36.5):
Calculated “Envelope” of
intensity intensity function f sin1b>22 2
I = I0 cos2 c d (two slits of finite width) (36.12)
0
u 2 b>2
where, as before,
(d) Photograph of the pattern calculated in (c)
2pd 2pa
f = sin u b = sin u
l l
In Fig. 36.12c, every fourth interference maximum at the sides is missing because
these interference maxima 1mi = {4, {8, c2 coincide with diffraction minima
1md = {1, {2, c2. This can also be seen in Fig. 36.12d, which is a photograph of an
For d = 4a, every fourth interference maximum actual pattern with d = 4a. You should be able to convince yourself that there will be
at the sides (mi = {4, {8, ...) is missing. “missing” maxima whenever d is an integer multiple of a.
 36.4 Multiple Slits    1195

Figures 36.12c and 36.12d show that as you move away from the central bright maxi-
mum of the two-slit pattern, the intensity of the maxima decreases. This is a result of
the single-slit modulating pattern shown in Fig. 36.12a; mathematically, the decrease in
intensity arises from the factor 1b>22 2 in the denominator of Eq. (36.12). You can also see
this decrease in Fig. 35.6 (Section 35.2). The narrower the slits, the broader the single-slit
pattern (as in Fig. 36.10) and the slower the decrease in intensity from one interference
maximum to the next.
Shall we call the pattern in Fig. 36.12d interference or diffraction? It’s really both,
since it results from the superposition of waves coming from various parts of the two
apertures.

Several Slits
Next let’s consider patterns produced by several very narrow slits. As we’ll see, systems Figure 36.13 Multiple-slit diffraction.
of narrow slits are of tremendous practical importance in spectroscopy, the determina- Here a lens is used to give a Fraunhofer
pattern on a nearby screen, as in Fig. 36.4d.
tion of the particular wavelengths of light coming from a source. Assume that each slit
is narrow in comparison to the wavelength, so its diffraction pattern spreads out nearly
uniformly. Figure 36.13 shows an array of eight narrow slits, with distance d between
adjacent slits. Constructive interference occurs for rays at angle u to the normal that ar-
rive at point P with a path difference between adjacent slits equal to an integer number
of wavelengths: P

d sin u = ml 1m = 0, {1, {2, c2 u

u
This means that reinforcement occurs when the phase difference f at P for light from
adjacent slits is an integer multiple of 2p. That is, the maxima in the pattern occur at the
same positions as for two slits with the same spacing. u
d
What happens between the maxima is different with multiple slits, however. In the two-
slit pattern, there is exactly one intensity minimum located midway between each pair d sin u
of maxima, corresponding to angles for which the phase difference between waves from Maxima occur where the path difference for
adjacent slits is a whole number of wavelengths:
the two sources is p, 3p, 5p, and so on. In the eight-slit pattern these are also minima
d sin u = ml.
because the light from adjacent slits cancels out in pairs, corresponding to the phasor
diagram in Fig. 36.14a. But these are not the only minima in the eight-slit pattern. For
example, when the phase difference f from adjacent sources is p>4, the phasor diagram
is as shown in Fig. 36.14b; the total (resultant) phasor is zero, and the intensity is zero.
When f = p>2, we get the phasor diagram of Fig. 36.14c, and again both the total phasor
and the intensity are zero. More generally, the intensity with eight slits is zero whenever
f is an integer multiple of p>4, except when f is a multiple of 2p. Thus there are seven
minima for every maximum.

Figure 36.14 Phasor diagrams for light passing through eight narrow slits. Intensity maxima occur
when the phase difference f = 0, 2p, 4p, c. Between the maxima at f = 0 and f = 2p are
seven minima, corresponding to f = p>4, p>2, 3p>4, p, 5p>4, 3p>2, and 7p>4. Can you draw
phasor diagrams for the other minima?
p p
(a) Phasor diagram for f = p (b) Phasor diagram for f = (c) Phasor diagram for f =
4 2

p
f = p = 180° f = = 90°
2

p
f = = 45°
4
1196    CHAPTER 36 Diffraction

Figure 36.15 Interference patterns for N Figure 36.15b shows the result of a detailed calculation of the eight-slit pattern. The
equally spaced, very narrow slits. (a) Two large maxima, called principal maxima, are in the same positions as for the two-slit
slits. (b) Eight slits. (c) Sixteen slits. The
pattern of Fig. 36.15a but are much narrower. If the phase difference f between adja-
vertical scales are different for each graph;
the maximum intensity is I0 for a single cent slits is slightly different from a multiple of 2p, the waves from slits 1 and 2 will
slit and N 2I0 for N slits. The width of each be only a little out of phase; however, the phase difference between slits 1 and 3 will be
peak is proportional to 1>N. greater, that between slits 1 and 4 will be greater still, and so on. This leads to a partial
(a) N = 2: two slits produce one minimum cancellation for angles that are only slightly different from the angle for a maximum,
between adjacent maxima. giving the narrow maxima in Fig. 36.15b. The maxima are even narrower with 16 slits
I (Fig. 36.15c).
4I0 You should show that when there are N slits, there are 1N - 12 minima between each
pair of principal maxima and a minimum occurs whenever f is an integer multiple of
2p>N (except when f is an integer multiple of 2p, which gives a principal maximum).
There are small secondary intensity maxima between the minima; these become smaller
in comparison to the principal maxima as N increases. The greater the value of N, the
narrower the principal maxima become. From an energy standpoint the total power in the
entire pattern is proportional to N. The height of each principal maximum is proportional
m = -1 m = 0 m = 1
u to N 2, so from energy conservation the width of each principal maximum must be propor-
tional to 1>N. In the next section we’ll see why the details of the multiple-slit pattern are
(b) N = 8: eight slits produce taller, narrower of great practical importance.
maxima in the same locations, separated by
seven minima.
TEST YOUR UNDERSTANDING OF SECTION 36.4 Suppose two slits, each of width a, are
I
separated by a distance d = 2.5a. Are there any missing maxima in the interference pattern pro-
64I0 duced by these slits? If so, which are missing? If not, why not?

ANSWER
because it coincides with the fourth diffraction minimum), and so on.
second diffraction minimum), mi = {10 and md = {4 (the tenth interference maximum is missing
for mi = {5 and md = {2 (the fifth interference maximum is missing because it coincides with the
d = 2.5a, we can combine these two conditions into the relationship mi = 2.5md . This is satisfied
an interference maximum) and a sin u = mdl (the condition for a diffraction minimum). Substituting
u ❙ yes; m i = t5, t10, N A “missing maximum” satisfies both d sin u = mil (the condition for
m = -1 m = 0 m = 1

(c) N = 16: with 16 slits, the maxima are even

taller and narrower, with more intervening
minima.
I 36.5 THE DIFFRACTION GRATING
256I0
We have just seen that increasing the number of slits in an interference experiment
(while keeping the spacing of adjacent slits constant) gives interference patterns in which
the maxima are in the same positions, but progressively narrower, than with two slits.
Because these maxima are so narrow, their angular position, and hence the wavelength,
can be measured to very high precision. As we’ll see, this effect has many important
applications.
m = -1 m = 0 m = 1
u An array of a large number of parallel slits, all with the same width a and spaced equal
distances d between centers, is called a diffraction grating. The first one was constructed
by Fraunhofer using fine wires. Gratings can be made by using a diamond point to scratch
many equally spaced grooves on a glass or metal surface, or by photographic reduction of
a pattern of black and white stripes on paper. For a grating, what we have been calling slits
are often called rulings or lines.
In Fig. 36.16, GG′ is a cross section of a transmission grating; the slits are
­p erpendicular to the plane of the page, and an interference pattern is formed by
the light that is transmitted through the slits. The diagram shows only six slits;
an ­actual grating may contain several thousand. The spacing d between centers of
­adjacent slits is called the grating spacing. A plane monochromatic wave is incident
normally on the grating from the left side. We assume far-field (Fraunhofer) con-
ditions; that is, the pattern is formed on a screen that is far enough away that all
rays emerging from the grating and going to a particular point on the screen can be
­considered to be parallel.
 36.5 The Diffraction Grating    1197

We found in Section 36.4 that the principal intensity maxima with multiple slits occur Figure 36.16 A portion of a transmission
in the same directions as for the two-slit pattern. These are the directions for which the diffraction grating. The separation be-
tween the centers of adjacent slits is d.
path difference for adjacent slits is an integer number of wavelengths. So the positions of
the maxima are once again G

Distance between slits Wavelength

Intensity maxima,
multiple slits:
d sin u = ml 1m = 0, {1, {2, c2 (36.13) d

Angle of line from center of slit array to mth bright region on screen
d

The intensity patterns for two, eight, and 16 slits displayed in Fig. 36.15 show the progres- d
sive increase in sharpness of the maxima as the number of slits increases.
When a grating containing hundreds or thousands of slits is illuminated by a beam of d
parallel rays of monochromatic light, the pattern is a series of very sharp lines at angles de-
termined by Eq. (36.13). The m = {1 lines are called the first-order lines, the m = {2 d
u
lines the second-order lines, and so on. If the grating is illuminated by white light with
a continuous distribution of wavelengths, each value of m corresponds to a continuous
spectrum in the pattern. The angle for each wavelength is determined by Eq. (36.13); for a G′
given value of m, long wavelengths (the red end of the spectrum) lie at larger angles (that
is, are deviated more from the straight-ahead direction) than do the shorter wavelengths at
the violet end of the spectrum.
As Eq. (36.13) shows, the sines of the deviation angles of the maxima are propor-
tional to the ratio l>d. For substantial deviation to occur, the grating spacing d should Figure 36.17 Microscopic pits on the
be of the same order of magnitude as the wavelength l. Gratings for use with visible surface of this DVD act as a reflec-
tion grating, splitting white light into its
light (l from 400 to 700 nm) usually have about 1000 slits per millimeter; the value ­component colors.
of d is the reciprocal of the number of slits per unit length, so d is of the order of
1
1000 mm = 1000 nm.
In a reflection grating, the array of equally spaced slits shown in Fig. 36.16 is re-
placed by an array of equally spaced ridges or grooves on a reflective screen. The re-
flected light has maximum intensity at angles where the phase difference between light
waves reflected from adjacent ridges or grooves is an integer multiple of 2p. If light of
wavelength l is incident normally on a reflection grating with a spacing d between ad-
jacent ridges or grooves, the reflected angles at which intensity maxima occur are given
by Eq. (36.13).
The rainbow-colored reflections from the surface of a DVD are a reflection-grating
effect (Fig. 36.17). The “grooves” are tiny pits 0.12 mm deep in the surface of the disc,
with a uniform radial spacing of 0.74 mm = 740 nm. Information is coded on the DVD
by varying the length of the pits. The reflection-grating aspect of the disc is merely an
aesthetic side benefit.

EXAMPLE 36.4 Width of a grating spectrum WITH ARIATION PROBLEMS

The wavelengths of the visible spectrum are approximately 380 nm We solve Eq. (36.13) for u:
(violet) to 750 nm (red). (a) Find the angular limits of the first-order vis-
ml
ible spectrum produced by a plane grating with 600 slits per millimeter u = arcsin
when white light falls normally on the grating. (b) Do the first-order d
and second-order spectra overlap? What about the second-order and Then for m = 1, the angular deviations uv1 and ur1 for violet and red
third-order spectra? Do your answers depend on the grating spacing? light, respectively, are

IDENTIFY and SET UP We must find the angles spanned by the visible 380 * 10-9 m
uv1 = arcsina b = 13.2°
spectrum in the first-, second-, and third-order spectra. These corre- 1.67 * 10-6 m
spond to m = 1, 2, and 3 in Eq. (36.13).
750 * 10-9 m
EXECUTE (a) The grating spacing is ur1 = arcsina b = 26.7°
1.67 * 10-6 m
1 That is, the first-order visible spectrum appears with deflection angles
d = = 1.67 * 10-6 m
600 slits>mm from uv1 = 13.2° (violet) to ur1 = 26.7° (red).
Continued
1198    CHAPTER 36 Diffraction

(b) With m = 2 and m = 3, our equation u = arcsin1ml>d2 for reaches u = 90° = arcsin112 at a wavelength shorter than 750 nm; you
380 mm violet light yields should be able to show that this happens for l = 557 nm. Hence the
first-order spectrum (from 13.2° to 26.7°) does not overlap with the
21380 * 10-9 m2 ­second-order spectrum, but the second- and third-order spectra do over-
uv2 = arcsina b = 27.1°
1.67 * 10-6 m lap. You can convince yourself that this is true for any value of the grat-
ing spacing d.
31380 * 10-9 m2
uv3 = arcsina b = 43.0°
1.67 * 10-6 m EVALUATE The fundamental reason the first-order and second-order
visible spectra don’t overlap is that the human eye is sensitive to only a
For 750 nm red light, this same equation gives
narrow range of wavelengths. Can you show that if the eye could detect
21750 * 10-9 m2 wavelengths from 380 nm to 900 nm (in the near-infrared range), the
ur2 = arcsina b = 63.9° first and second orders would overlap?
1.67 * 10-6 m
31750 * 10-9 m2 KEYCONCEPT In the interference pattern produced by a transmis-
ur3 = arcsina -6
b = arcsin11.352 = undefined sion or reflection grating, the intensity maxima occur at angles that are
1.67 * 10 m determined by the grating spacing and the wavelength of the light used.
Hence the second-order spectrum extends from 27.1° to 63.9° and the These angles are the same as in the interference pattern produced in the
third-order spectrum extends from 43.0° to 90° (the largest possible val- double-slit experiment; the difference is that the maxima are much nar-
ue of u). The undefined value of ur3 means that the third-order s­ pectrum rower for a grating.

Figure 36.18 (a) A visible-light Grating Spectrographs

p­ hotograph of the sun. (b) Sunlight is
dispersed into a spectrum by a diffraction Diffraction gratings are widely used to measure the spectrum of light emitted by a source,
grating. Specific wavelengths are absorbed a process called spectroscopy or spectrometry. Light incident on a grating of known
as sunlight passes through the sun’s atmo- ­spacing is dispersed into a spectrum. The angles of deviation of the maxima are then
sphere, leaving dark lines in the spectrum. measured, and Eq. (36.13) is used to compute the wavelength. With a grating that has
(a) many slits, very sharp maxima are produced, and the angle of deviation (and hence the
wavelength) can be measured very precisely.
An important application of this technique is to astronomy. As light generated
within the sun passes through the sun’s atmosphere, certain wavelengths are selectively
­absorbed. The result is that the spectrum of sunlight produced by a diffraction grat-
ing has dark absorption lines (Fig. 36.18). Experiments in the laboratory show that
different types of atoms and ions absorb light of different wavelengths. By comparing
these laboratory results with the wavelengths of absorption lines in the spectrum of sun-
light, astronomers can deduce the chemical composition of the sun’s atmosphere. The
same technique is used to make chemical assays of galaxies that are millions of light-
years away.
Figure 36.19 shows one design for a grating spectrograph used in a­ stronomy. A trans-
mission grating is used in the figure; in other setups, a reflection grating is used. In older
designs a prism was used rather than a grating, and a spectrum was formed by dispersion
(b) (see Section 33.4) rather than diffraction. However, there is no simple relationship be-
tween wavelength and angle of deviation for a prism, prisms absorb some of the light that
passes through them, and they are less effective for many nonvisible w ­ avelengths that are
important in astronomy. For these and other reasons, gratings are preferred in precision
applications.

Resolution of a Grating Spectrograph

In spectroscopy it is often important to distinguish slightly differing wavelengths. The
minimum wavelength difference ∆l that can be distinguished by a spectrograph is
­described by the chromatic resolving power R, defined as

l
R = (chromatic resolving power)(36.14)
∆l
 36.5 The Diffraction Grating    1199

1 Light from telescope Figure 36.19 A schematic diagram of a

is sent along fiber-optic diffraction-grating spectrograph for use
cables (not shown) and in astronomy. Note that the light does not
emerges here. strike the grating normal to its surface, so
6 An electronic detector the intensity maxima are given by a some-
(like the one in what different expression than Eq. (36.13).
a digital camera)
records the spectrum.

2 Light strikes concave

mirror and emerges as a
beam of parallel rays.

3 Light passes through diffraction grating.

4 Lenses direct 5 Concave mirror

diffracted light onto a reflects light to a
second concave mirror. focus.
BIO APPLICATION Detecting DNA
with Diffraction Diffraction gratings are
used in a common piece of laboratory
As an example, when sodium atoms are heated, they emit strongly at the yellow wave- equipment known as a spectrophotometer.
Light shining across a diffraction grating is
lengths 589.00 nm and 589.59 nm. A spectrograph that can barely distinguish these
dispersed into its component wavelengths.
two lines in the spectrum (called the sodium doublet) has a chromatic resolving power A slit is used to block all but a very narrow
R = 1589.00 nm2>10.59 nm2 = 1000. (You can see these wavelengths when boiling range of wavelengths, producing a beam
water with table salt in it on a gas range. If the water boils over onto the flame, dissolved of almost perfectly monochromatic light.
sodium from the salt emits a burst of yellow light.) The instrument then measures how much
of that light is absorbed by a solution of
We can derive an expression for the resolving power of a diffraction grating used in
biological molecules. For example, the
a spectrograph. Two different wavelengths give diffraction maxima at slightly different sample tube shown here contains a solu-
angles. As a reasonable (though arbitrary) criterion, let’s assume that we can distinguish tion of DNA, which is transparent to visible
them as two separate peaks if the maximum of one coincides with the first minimum of light but which strongly absorbs ultraviolet
the other. light with a wavelength of exactly 260 nm.
Therefore, by illuminating the sample with
From our discussion in Section 36.4 the mth@order maximum occurs when the phase
260 nm light and measuring the amount
difference f for adjacent slits is f = 2pm. The first minimum beside that maximum oc- absorbed, we can determine the concentra-
curs when f = 2pm + 2p>N, where N is the number of slits. The phase difference is tion of DNA in the solution.
also given by f = 12pd sin u2>l, so the angular interval du corresponding to a small in-
crement df in the phase shift can be obtained from the differential of this equation:
2pd cos u du
df =
l
When df = 2p>N, this corresponds to the angular interval du between a maximum and
the first adjacent minimum. Thus du is given by

2p 2pd cos u du l
= or d cos u du =
N l N

Now we need to find the angular spacing du between maxima for two slightly different
wavelengths. The positions of these maxima are given by d sin u = ml, and the differen-
tial of this equation gives
d cos u du = m dl

According to our criterion, the limit or resolution is reached when these two angular spac-    CAUTION    Watch out for different uses
ings are equal. Equating the two expressions for the quantity 1d cos u du2, we find of the symbol d Don’t confuse the slit
spacing d with the differential “d” in the
l l angular interval du or in the phase shift
= m dl and = Nm
N dl ­increment df! ❙
1200    CHAPTER 36 Diffraction

If ∆l is small, we can replace dl by ∆l, and the resolving power R is

l
R = = Nm (36.15)
∆l
The greater the number of slits N, the better the resolution; also, the higher the order m of
the diffraction-pattern maximum that we use, the better the resolution.

TEST YOUR UNDERSTANDING OF SECTION 36.5 What minimum number of slits would be
required in a grating to resolve the sodium doublet in the fourth order? (i) 250; (ii) 400; (iii) 1000;
(iv) 4000.

ANSWER
the shapes and spacings of the slits.)
­nature of our criterion for resolution and because real gratings always have slight imperfections in

order 1m = 12 we need N = 1000 slits, but in the fourth order 1m = 42 we need only
N = R>m = 1000>4 = 250 slits. (These numbers are only approximate because of the arbitrary

❙ (i) As described in the text, the resolving power needed is R = Nm = 1000. In the first

36.6 X-RAY DIFFRACTION

X rays were discovered by Wilhelm Röntgen (1845–1923) in 1895, and early experiments
suggested that they were electromagnetic waves with wavelengths of the order of 10-10 m.
At about the same time, the idea began to emerge that in a crystalline solid the atoms are
arranged in a regular repeating pattern, with spacing between adjacent atoms also of the
order of 10-10 m. Putting these two ideas together, Max von Laue (1879–1960) proposed
in 1912 that a crystal might serve as a kind of three-dimensional diffraction grating for x
rays. That is, a beam of x rays might be scattered (that is, absorbed and re-emitted) by the
individual atoms in a crystal, and the scattered waves might interfere just like waves from
a diffraction grating.
The first x-ray diffraction experiments were performed in 1912 by Friedrich,
Knipping, and von Laue, using the experimental setup shown in Fig. 36.20a. The scat-
tered x rays did form an interference pattern, which they recorded on photographic film.
Figure 36.20b is a photograph of such a pattern. These experiments verified that x rays
are waves, or at least have wavelike properties, and also that the atoms in a crystal are
arranged in a regular pattern (Fig. 36.21). Since that time, x-ray diffraction has proved to
be an invaluable research tool, both for measuring x-ray wavelengths and for studying the
structure of crystals and complex molecules.

Figure 36.20 (a) An x-ray diffraction experiment. (b) Diffraction pattern (or Laue pattern) formed
by directing a beam of x rays at a thin section of beryl crystal.
(a) Basic setup for x-ray diffraction (b) Laue diffraction pattern for a thin section
Some x rays are scattered as they pass of beryl crystal
through the crystal, forming an interference
pattern on the film. (Most of the x rays pass
straight through the crystal.)

Thin
Lead crystal
screen
X-ray
tube
X-ray beam

Detector
sensitive
to x rays
 36.6 X-Ray Diffraction    1201

A Simple Model of X-Ray Diffraction Figure 36.21 Model of the arrange-

ment of ions in a crystal of NaCl (table
To better understand x-ray diffraction, we consider first a two-dimensional scattering situ- salt). The spacing of adjacent atoms is
ation, as shown in Fig. 36.22a, in which a plane wave is incident on a rectangular array 0.282 nm. (The electron clouds of the
of scattering centers. The situation might be a ripple tank with an array of small posts or atoms ­actually overlap slightly.)
x rays incident on an array of atoms. In the case of electromagnetic waves, the wave in-
duces an oscillating electric dipole moment in each scatterer. These dipoles act like little Chloride
antennas, emitting scattered waves. The resulting interference pattern is the superposition ions
of all these scattered waves. The situation is different from that with a diffraction grat-
ing, in which the waves from all the slits are emitted in phase (for a plane wave at normal
incidence). Here the scattered waves are not all in phase because their distances from the
source are different. To compute the interference pattern, we have to consider the total
path differences for the scattered waves, including the distances from source to scatterer
and from scatterer to observer.
As Fig. 36.22b shows, the path length from source to observer is the same for all the
scatterers in a single row if the two angles ua and ur are equal. Scattered radiation from
adjacent rows is also in phase if the path difference for adjacent rows is an integer number
of wavelengths. Figure 36.22c shows that this path difference is 2d sin u, where u is the
common value of ua and ur . Therefore the conditions for radiation from the entire array to Sodium
reach the observer in phase are (1) the angle of incidence must equal the angle of scatter- ions
ing and (2) the path difference for adjacent rows must equal ml, where m is an integer. We
can express the second condition, called the Bragg condition in honor of x-ray diffraction
pioneers Sir William Bragg and his son Laurence Bragg, as

Bragg condition Distance between

for constructive    CAUTION    Scattering from an array In
adjacent rows in array Wavelength
interference Eq. (36.16) the angle u is measured with
from an array: 2d sin u = ml 1m = 1, 2, 3, c2 (36.16) respect to the surface of the crystal rather
Angle of line from surface of array to mth bright region on screen than with respect to the normal to the
plane of an array of slits or a grating. Also,
note that the path difference in Eq. (36.16)
In directions for which Eq. (36.16) is satisfied, we see a strong maximum in the inter-
is 2d sin u, not d sin u as in Eq. (36.13) for a
ference pattern. We can describe this interference in terms of reflections of the wave from
diffraction grating. ❙
the horizontal rows of scatterers in Fig. 36.22a. Strong reflection (constructive interfer-
ence) occurs at angles such that the incident and scattered angles are equal and Eq. (36.16)
is satisfied. Since sin u can never be greater than 1, Eq. (36.16) says that to have con-
structive interference the quantity ml must be less than 2d and so l must be less than
2d>m. For example, the value of d in an NaCl crystal (see Fig. 36.21) is only 0.282 nm.
Hence to have the mth@order maximum present in the diffraction pattern, l must be
less than 210.282 nm2>m; that is, l 6 0.564 nm for m = 1, l 6 0.282 nm for m = 2,
l 6 0.188 nm for m = 3, and so on. These are all x-ray wavelengths (see Fig. 32.4),
which is why x rays are used for studying crystal structure.

Figure 36.22 A two-dimensional model of scattering from a rectangular array. The distance between
adjacent atoms in a horizontal row is a; the distance between adjacent rows is d. The angles in
(b) are measured from the surface of the array, not from its normal.
(a) Scattering of waves from a rectangular array (b) Scattering from adjacent atoms in a row (c) Scattering from atoms in adjacent rows
Interference from adjacent atoms in a row is Interference from atoms in adjacent rows is
Incident plane waves constructive when the path lengths a cos ua constructive when the path difference
and a cos ur are equal, so that the angle of 2d sin u is an integer number of
incidence ua equals the angle of reflection wavelengths, as in Eq. (36.16).
(scattering) ur .

a cos ua a cos ur
Scatterers (e.g., atoms)
d uu
ur ua d
ua ur
a a d sin u d sin u
1202    CHAPTER 36 Diffraction

Figure 36.23 A cubic crystal and two (a) Spacing of planes is d = a>12. (b) Spacing of planes is d = a>13.
different families of crystal planes. There
are also three sets of planes parallel to the
cube faces, with spacing a. a

   CAUTION    Bragg reflection is really

Bragg interference While we are using a
the term reflection, remember that we are
dealing with an interference effect. The
reflections from various planes are closely
analogous to interference effects in thin
films (see Section 35.4). ❙ We can extend this discussion to a three-dimensional array by considering planes of
scatterers instead of rows. Figure 36.23 shows two different sets of parallel planes that
pass through all the scatterers. Waves from all the scatterers in a given plane interfere
Figure 36.24 The British scientist constructively if the angles of incidence and scattering are equal. There is also construc-
Rosalind Franklin made this groundbreak- tive interference between planes when Eq. (36.16) is satisfied, where d is now the distance
ing x-ray diffraction image of DNA in
1953. The dark bands arranged in a cross between adjacent planes. Because there are many different sets of parallel planes, there
provided the first evidence of the helical are also many values of d and many sets of angles that give constructive interference for
structure of the DNA molecule. the whole crystal lattice. This phenomenon is called Bragg reflection.
As Fig. 36.20b shows, in x-ray diffraction there is nearly complete cancellation in all
but certain very specific directions in which constructive interference occurs and forms
bright spots. Such a pattern is usually called an x-ray diffraction pattern, although inter-
ference pattern might be more appropriate.
We can determine the wavelength of x rays by examining the diffraction pattern for
a crystal of known structure and known spacing between atoms, just as we determined
wavelengths of visible light by measuring patterns from slits or gratings. (The spacing
between atoms in simple crystals of known structure, such as sodium chloride, can be
found from the density of the crystal and Avogadro’s number.) Then, once we know the
x-ray wavelength, we can use x-ray diffraction to explore the structure and determine the
spacing between atoms in crystals with unknown structure.
X-ray diffraction is by far the most important experimental tool in the investigation of
crystal structure of solids. X-ray diffraction also plays an important role in studies of the
structures of liquids and of organic molecules. It has been one of the chief experimental
techniques in working out the double-helix structure of DNA (Fig. 36.24) and subsequent
advances in molecular genetics.

EXAMPLE 36.5 X-ray diffraction WITH ARIATION PROBLEMS

You direct a beam of 0.154 nm x rays at certain planes of a silicon crystal. (b) To calculate other angles, we solve Eq. (36.16) for sin u:
As you increase the angle of incidence of the beam from zero, the first
ml 0.154 nm
strong interference maximum occurs when the beam makes an angle of sin u = = m = m10.5662
34.5° with the planes. (a) How far apart are the planes? (b) Will you find 2d 210.136 nm2
other interference maxima from these planes at larger angles of incidence? Values of m of 2 or greater give values of sin u greater than unity, which
IDENTIFY and SET UP This problem involves Bragg reflection of x rays is impossible. Hence there are no other angles for interference maxima
from the planes of a crystal. In part (a) we use the Bragg condition, for this particular set of crystal planes.
Eq. (36.16), to find the distance d between adjacent planes from the EVALUATE Our result in part (b) shows that there would be a second
known wavelength l = 0.154 nm and angle of incidence u = 34.5° interference maximum if the quantity 2l>2d = l>d were less than 1.
for the m = 1 interference maximum. Given the value of d, we use the This would be the case if the wavelength of the x rays were less than
Bragg condition again in part (b) to find the values of u for interference d = 0.136 nm. How short would the wavelength need to be to have
maxima corresponding to other values of m. three interference maxima?
EXECUTE (a) We solve Eq. (36.16) for d and set m = 1: KEYCONCEPT In x-ray diffraction, waves that pass through or re-
flect from a crystal are scattered by rows of atoms within the crystal.
ml 11210.154 nm2
d = = = 0.136 nm The pattern of scattered light has maxima at points where waves that
2 sin u 2 sin 34.5° interact with different rows arrive in phase and interfere constructively
This is the distance between adjacent planes. [Eq. (36.16)].
 36.7 Circular Apertures and Resolving Power     1203

TEST YOUR UNDERSTANDING OF SECTION 36.6 You are doing an x-ray diffraction exper-
iment with a crystal in which the atomic planes are 0.200 nm apart. You are using x rays of wave-
length 0.0900 nm. What is the highest-order maximum present in the diffraction pattern? (i) Third;
(ii) fourth; (iii) fifth; (iv) sixth; (v) seventh.

ANSWER
maxima do not appear.
integer, the highest-order maximum in the pattern is m = 4 (fourth order). The m = 5, 6, 7, c

m = 12d sin u2>l. The sine function can never be greater than 1, so the largest value of m in the
­pattern can be no greater than 2d>l = 210.200 nm2>10.0900 nm2 = 4.44. Since m must be an

❙ (ii) The angular position of the mth maximum is given by Eq. (36.16), 2d sin u = ml. This gives

36.7 CIRCULAR APERTURES AND RESOLVING POWER

We have studied in detail the diffraction patterns formed by long, thin slits or arrays of
APPLICATION Bigger Telescope,
slits. But an aperture of any shape forms a diffraction pattern. The diffraction pattern Better Resolving Power The large ap-
formed by a circular aperture is of special interest because of its role in limiting how well erture ­diameter of very large telescopes
an optical instrument can resolve fine details. In principle, we could compute the intensity minimizes diffraction effects. The effective
at any point P in the diffraction pattern by dividing the area of the aperture into small ele- ­diameter of a telescope can be increased
by using arrays of smaller telescopes. The
ments, finding the resulting wave amplitude and phase at P, and then integrating over the
Very Large Array (VLA) in New Mexico is
aperture area to find the resultant amplitude and intensity at P. In practice, the integration a collection of 27 radio telescopes, each
cannot be carried out in terms of elementary functions. We’ll simply describe the pattern 25 m in diameter, that can be spread out
and quote a few relevant numbers. in a Y-shaped arrangement 36 km across.
The diffraction pattern formed by a circular aperture consists of a central bright spot Hence the effective aperture diameter is
36 km, giving the VLA a limit of resolu-
surrounded by a series of bright and dark rings, as Fig. 36.25 shows. We can describe the
tion of 5 * 10-8 rad at a radio wavelength
pattern in terms of the angle u, representing the angular radius of each ring. The angular of 1.5 cm. If your eye had this resolving
radius u1 of the first dark ring is given by power, you could read the “20>20” line
on an eye chart more than 30 km away!
Angular radius of first dark ring = angular radius of Airy disk
Diffraction by a l Wavelength
circular aperture: sin u1 = 1.22 (36.17)
D Aperture diameter

The angular radii of the next two dark rings are given by

l l
sin u2 = 2.23 sin u3 = 3.24 (36.18)
D D
The central bright spot is called the Airy disk, in honor of Sir George Airy (1801–1892),
who first derived the expression for the intensity in the pattern. The angular radius of the
Airy disk is that of the first dark ring, given by Eq. (36.17). The angular radii of the first
three bright rings outside the Airy disk are

l l l
sin u = 1.63 , 2.68 , 3.70 (36.19)
D D D

Figure 36.25 Diffraction pattern formed

by a circular aperture of diameter D. The
pattern consists of a central bright spot
u1 is the angle between the center of and alternating dark and bright rings. The
the pattern and the first minimum. angular radius u1 of the first dark ring
is shown. (This diagram is not drawn to
Airy scale.)
disk
u1

D
1204    CHAPTER 36 Diffraction

Figure 36.26 Photograph of the diffrac- The intensities in the bright rings drop off very quickly with increasing angle. When
tion pattern formed by a circular aperture. D is much larger than the wavelength l, the usual case for optical instruments, the peak
intensity in the first ring is only 1.7% of the value at the center of the Airy disk, and the
Airy disk
peak intensity of the second ring is only 0.4%. Most (85%) of the light energy falls within
the Airy disk. Figure 36.26 shows a diffraction pattern from a circular aperture.

Diffraction and Image Formation

Diffraction has far-reaching implications for image formation by lenses and mirrors. In
our study of optical instruments in Chapter 34 we assumed that a lens with focal length
f focuses a parallel beam (plane wave) to a point at a distance f from the lens. We now
see that what we get is not a point but the diffraction pattern just described. If we have
two point objects, their images are not two points but two diffraction patterns. When the
objects are close together, their diffraction patterns overlap; if they are close enough, their
patterns overlap almost completely and cannot be distinguished. The effect is shown in
Fig. 36.27, which presents the patterns for four very small “point” sources of light. In
Fig. 36.27a the image of source 1 is well separated from the others, but the images of
sources 3 and 4 have merged. In Fig. 36.27b, with a larger aperture diameter and hence
smaller Airy disks, the images of sources 3 and 4 are better resolved. In Fig. 36.27c, with
a still larger aperture, they are well resolved.
Figure 36.27 These images show the A widely used criterion for resolution of two point objects, proposed by the English
diffraction patterns produced when four
very small (“point”) sources of light are physicist Lord Rayleigh (1842–1919) and called Rayleigh’s criterion, is that the objects
viewed through a lens with a circular are just barely resolved (that is, distinguishable) if the center of one diffraction pattern
aperture placed in front of the lens. The coincides with the first minimum of the other. In that case the angular separation of the
photographs were made with a circular image centers is given by Eq. (36.17). The angular separation of the objects is the same as
­aperture in front of the lens. (a) The aper- that of the images made by a telescope, microscope, or other optical device. So two point
ture is so small that the patterns of sources
3 and 4 overlap and are barely resolved by objects are barely resolved when their angular separation is given by Eq. (36.17).
Rayleigh’s criterion. Increasing the size of The minimum separation of two objects that can just be resolved by an optical instru-
the aperture decreases the size of the dif- ment is called the limit of resolution of the instrument. The smaller the limit of resolu-
fraction patterns, as shown in (b) and (c). tion, the greater the resolution, or resolving power, of the instrument. Diffraction sets the
(a) Small aperture ultimate limits on resolution of lenses. Geometric optics may make it seem that we can
make images as large as we like. Eventually, though, we always reach a point at which
the image becomes larger but does not gain in detail. The images in Fig. 36.27 would not
become sharper if enlarged.

   CAUTION    Resolving power vs. chromatic resolving power Don’t confuse the resolving power of
an optical instrument with the chromatic resolving power of a grating (Sec­tion 36.5). Resolving
1 2 3 4 power refers to the ability to distinguish the images of objects that appear close to each other, when
looking either through an optical instrument or at a photograph made with the instrument.
Chromatic resolving power describes how well different wavelengths can be distinguished in a
(b) Medium aperture spectrum formed by a diffraction grating. ❙

Rayleigh’s criterion combined with Eq. (36.17) shows that resolution (resolving
power) improves with larger diameter; it also improves with shorter wavelengths. ?
Ultraviolet microscopes have higher resolution than visible-light microscopes. In
electron microscopes the resolution is limited by the wavelengths associated with the
electrons, which have wavelike aspects (to be discussed further in Chapter 39). These
1 2 3 4 wavelengths can be made 100,000 times smaller than wavelengths of visible light, with
a corresponding gain in resolution. Resolving power also explains the difference in
(c) Large aperture storage capacity between DVDs (introduced in 1995) and Blu-ray discs (introduced in
2003). Information is stored in both of these in a series of tiny pits. In order not to lose
information in the scanning process, the scanning optics must be able to resolve two
adjacent pits so that they do not seem to blend into a single pit (see sources 3 and 4 in
Fig. 36.27). The blue scanning laser used in a Blu-ray player has a shorter wavelength
(405 nm) and hence better resolving power than the 650 nm red laser in a DVD player.
1 2 3 4 Hence pits can be spaced closer together in a Blu-ray disc than in a DVD, and more in-
formation can be stored on a disc of the same size (50 gigabytes on a Blu-ray disc versus
4.7 gigabytes on a DVD).
 36.7 Circular Apertures and Resolving Power     1205

BIO APPLICATION The Airy Disk in an Eagle’s Eye Diffraction by the pupil
of an eye limits resolving power. In a human’s eye, the maximum pupil diameter
D is about 5 mm; in an eagle’s eye, D is about 9 mm. From Eq. (36.17) this
means that an eagle’s eye has superior resolution: A distant point source of light
produces an Airy disk on an eagle’s retina that is only about 59 the angular size of
the disk produced on the retina of the human eye. (If our eye produces an image
like Fig. 36.27b, an eagle’s eye produces one like Fig. 36.27c.) To record the fine
­details of this high-resolution image, the light-sensitive cones in an eagle’s retina
are smaller and more closely packed than those in a human retina.

EXAMPLE 36.6 Resolving power of a camera lens WITH ARIATION PROBLEMS

vA camera lens with focal length f = 50 mm and maximum aperture y

= 2.4 * 10-5 y = 2.2 * 10-4 m = 0.22 mm
f>2 forms an image of an object 9.0 m away. (a) If the resolution is lim- 9.0 m
ited by diffraction, what is the minimum distance between two points y′
on the object that are barely resolved? What is the corresponding dis- = 2.4 * 10-5 y′ = 1.2 * 10-3 mm
50 mm
tance between image points? (b) How does the situation change if the 1
lens is “stopped down” to f>16? Use l = 500 nm in both cases. = 0.0012 mm ≈ 800 mm

IDENTIFY and SET UP This example uses the ideas about resolving
(b) The aperture diameter is now 150 mm2>16, or one-eighth as
power, image formation by a lens (Section 34.4), and f@number (Section
large as before. The angular separation between barely resolved points
34.5). From Eq. (34.20), the f@number of a lens is its focal length f
is eight times as great, and the values of y and y′ are also eight times
divided by the aperture diameter D. We use this equation to determine
as great as before:
D and then use Eq. (36.17) (the Rayleigh criterion) to find the angu-
1
lar separation u between two barely resolved points on the object. We y = 1.8 mm y′ = 0.0096 mm = 100 mm
then use the geometry of image formation by a lens to determine the
distance y between those points and the distance y′ between the cor- Only the best camera lenses can approach this resolving power.
responding image points. EVALUATE Many photographers use the smallest possible aperture for
EXECUTE (a) The aperture diameter is D = f>1f@number2 = maximum sharpness, since lens aberrations cause light rays that are far
150 mm2>2 = 25 mm = 25 * 10-3 m. From Eq. (36.17) the angular from the optic axis to converge to a different image point than do rays
separation u of two object points that are barely resolved is near the axis. But as this example shows, diffraction effects become
more significant at small apertures. One cause of fuzzy images has to be
l 500 * 10-9 m balanced against another.
u ≈ sin u = 1.22 = 1.22 = 2.4 * 10-5 rad
D 25 * 10-3 m KEYCONCEPT Like waves that pass through a narrow slit, waves
We know from our thin-lens analysis in Section 34.4 that, apart that pass through a circular aperture undergo diffraction and spread out.
from sign, y>s = y′>s′ [see Eq. (34.14)]. Thus the angular separations This sets a limit on the ability of a lens or telescope to resolve objects
of the object points and the corresponding image points are both equal that are close to each other.
to u. Because the object distance s is much greater than the focal length
f = 50 mm, the image distance s′ is approximately equal to f. Thus

TEST YOUR UNDERSTANDING OF SECTION 36.7 You have been asked to compare four pro-
posals for telescopes to be placed in orbit above the blurring effects of the earth’s atmosphere. Rank
the proposed telescopes in order of their ability to resolve small details, from best to worst. (i) A radio
telescope 100 m in diameter observing at a wavelength of 21 cm; (ii) an optical telescope 2.0 m in diam-
eter observing at a wavelength of 500 nm; (iii) an ultraviolet telescope 1.0 m in diameter observing at a
wavelength of 100 nm; (iv) an infrared telescope 2.0 m in diameter observing at a wavelength of 10 mm.

(iv) 110 mm2>12.0 m2 = 11.0 * 10-5 m2>12.0 m2 = 5.0 * 10-6.

(iii) 1100 nm2>11.0 m2 = 11.0 * 10-7 m2>11.0 m2 = 1.0 * 10-7;
ANSWER
(ii) 1500 nm2>12.0 m2 = 15.0 * 10-7 m2>12.0 m2 = 2.5 * 10-7;
telescopes, this ratio is equal to (i) 121 cm2>1100 m2 = 10.21 m2>1100 m2 = 2.1 * 10-3;
value of the ratio l>D, the better the resolving power of a telescope of diameter D. For the four
❙ (iii), (ii), (iv), (i) Rayleigh’s criterion combined with Eq. (36.17) shows that the smaller the
1206    CHAPTER 36 Diffraction

Figure 36.28 (a) A hologram is the record on film of the interference pattern formed with light from the
coherent source and light scattered from the object. (b) Images are formed when light is projected through
the hologram. The observer sees the virtual image formed behind the hologram.
(a) Recording a hologram (b) Viewing the hologram

Beam
splitter Photographic Observer
Laser Mirror
film

Object Real
beam image
Reference
beam Transparent
positive
film

Mirror Object Mirror

Mirror

Virtual
Reconstruction image
beam

36.8 HOLOGRAPHY
Holography is a technique for recording and reproducing an image of an object through
the use of interference effects. Unlike the two-dimensional images recorded by an or-
dinary photograph or television system, a holographic image is truly three-dimensional.
Such an image can be viewed from different directions to reveal different sides and from
various distances to reveal changing perspective. If you had never seen a hologram, you
wouldn’t believe it was possible!
Figure 36.28a shows the basic procedure for making a hologram. We illuminate the ob-
ject to be holographed with monochromatic light, and we place a photographic film so that it
is struck by scattered light from the object and also by direct light from the source. In practice,
the light source must be a laser, for reasons we’ll discuss later. Interference between the direct
and scattered light forms a complex interference pattern that is recorded on the film.
To form the images, we simply project light through the developed film (Fig. 36.28b).
Two images are formed: a virtual image on the side of the film nearer the source and a
real image on the opposite side.

Holography and Interference Patterns

A complete analysis of holography is beyond our scope, but we can gain some insight into the
process by looking at how a single point is holographed and imaged. Consider the interfer-
ence pattern that is formed on a sheet of photographic negative film by the superposition of
an incident plane wave and a spherical wave, as shown in Fig. 36.29a. The spherical wave
originates at a point source P at a distance b 0 from the film; P may in fact be a small object
that scatters part of the incident plane wave. We assume that the two waves are monochro-
matic and coherent and that the phase relationship is such that constructive interference oc-
curs at point O on the diagram. Then constructive interference will also occur at any point
Q on the film that is farther from P than O is by an integer number of wavelengths. That is,
if bm - b 0 = ml, where m is an integer, then constructive interference occurs. The points
where this condition is satisfied form circles on the film centered at O, with radii rm given by

bm - b 0 = 2b02 + rm2 - b 0 = ml 1m = 1, 2, 3, c2 (36.20)

Solving this for rm2, we find

rm2 = l12mb 0 + m2l2
 36.8 Holography    1207

Figure 36.29 (a) Constructive interference of the plane and spherical waves occurs in the plane of the film at every
point Q for which the distance bm from P is greater than the distance b 0 from P to O by an integer ­number of wave-
lengths ml. For the point Q shown, m = 2. (b) When a plane wave strikes a transparent positive print of the developed
film, the diffracted wave consists of a wave converging to P′ and then diverging again and a diverging wave that
­appears to originate at P. These waves form the real and virtual images, respectively.
(a) (b)

Plane wave Spherical wave Plane wave

Q Q
bm
l
rm

P b0 O P O P′

Film Transparent positive print

Figure 36.30 Two views of the same

Ordinarily, b 0 is very much larger than l, so we ignore the second term in parentheses and obtain hologram seen from different angles.

rm = 22mlb 0 1m = 1, 2, 3, c2(36.21)

The interference pattern consists of a series of concentric bright circular fringes with radii
given by Eq. (36.21). Between these bright fringes are dark fringes.
Now we develop the film and make a transparent positive print, so the bright-fringe
areas have the greatest transparency on the film. Then we illuminate it with monochro-
matic plane-wave light of the same wavelength l that we used initially. In Fig. 36.29b,
consider a point P′ at a distance b 0 along the axis from the film. The centers of successive
bright fringes differ in their distances from P′ by an integer number of wavelengths, and
therefore a strong maximum in the diffracted wave occurs at P′. That is, light converges to
P′ and then diverges from it on the opposite side. Therefore P′ is a real image of point P.
This is not the entire diffracted wave, however. The interference of the wavelets that
spread out from all the transparent areas forms a second spherical wave that is diverging
rather than converging. When this wave is traced back behind the film in Fig. 36.29b, it ap-
pears to be spreading out from point P. Thus the total diffracted wave from the hologram
is a superposition of a spherical wave converging to form a real image at P′ and a spherical
wave that diverges as though it had come from the virtual image point P.
Because of the principle of superposition for waves, what is true for the imaging of a sin-
gle point is also true for the imaging of any number of points. The film records the super-
posed interference pattern from the various points, and when light is projected through the
film, the various image points are reproduced simultaneously. Thus the images of an ex-
tended object can be recorded and reproduced just as for a single point object. Figure 36.30
shows photographs of a holographic image from two different angles, showing the chang-
ing perspective in this three-dimensional image.
In making a hologram, we have to overcome two practical problems. First, the light used
must be coherent over distances that are large in comparison to the dimensions of the object
and its distance from the film. Ordinary light sources do not satisfy this requirement, for
reasons that we discussed in Section 35.1. Therefore laser light is essential for making a holo-
gram. (Ordinary white light can be used for viewing certain types of hologram, such as those
used on credit cards.) Second, extreme mechanical stability is needed. If any relative motion
of source, object, or film occurs during exposure, even by as much as a quarter of a wave-
length, the interference pattern on the film is blurred enough to prevent satisfactory image
formation. These obstacles are not insurmountable, however, and holography has become im-
portant in research, entertainment, and a wide variety of technological applications.
1208    CHAPTER 36 Diffraction

CHAPTER 36 SUMMARY
Fresnel and Fraunhofer diffraction: Diffraction occurs when Fresnel (near-field) Fraunhofer (far-
light passes through an aperture or around an edge. When the diffraction field) diffraction
source and the observer are so far away from the obstructing sur-
face that the outgoing rays can be considered parallel, it is called
Fraunhofer diffraction. When the source or the observer is rela-
tively close to the obstructing surface, it is Fresnel diffraction. P

Single-slit diffraction: Monochromatic light sent through a nar- ml u

sin u = I = 0.0083I0 m = 3
row slit of width a produces a diffraction pattern on a distant a (36.2) I = 0.0165I0
screen. Equation (36.2) gives the condition for destructive in- m = 2
1m = {1, {2, {3, c2 I = 0.0472I0 m = 1
terference (a dark fringe) at a point P in the pattern at angle u. O
Equation (36.7) gives the intensity in the pattern as a function sin3pa1sin u2>l4 2 m = -1
I = I0 u
of u. (See Examples 36.1–36.3.) I = I0 b r  (36.7) m = -2
pa1sin u2>l
m = -3

Diffraction gratings: A diffraction grating consists of a large d sin u = ml (36.13) I

256I0
number of thin parallel slits, spaced a distance d apart. The con- 1m = 0, {1, {2, {3, c2
dition for maximum intensity in the interference pattern is the N = 16
same as for the two-source pattern, but the maxima for the grat-
ing are very sharp and narrow. (See Example 36.4.) u
m = -1 m = 0 m = 1

X-ray diffraction: A crystal serves as a three-dimensional dif- 2d sin u = ml (36.16)

fraction grating for x rays with wavelengths of the same order 1m = 1, 2, 3, c2
of magnitude as the spacing between atoms in the crystal. For
a set of crystal planes spaced a distance d apart, constructive d
interference occurs when the angles of incidence and scattering u u
(measured from the crystal planes) are equal and when the Bragg
condition [Eq. (36.16)] is satisfied. (See Example 36.5.) d sin u d sin u

Circular apertures and resolving power: The diffraction pat- l

sin u1 = 1.22  (36.17)
tern from a circular aperture of diameter D consists of a central D
bright spot, called the Airy disk, and a series of concentric dark
and bright rings. Equation (36.17) gives the angular radius u1 Airy disk
of the first dark ring, equal to the angular size of the Airy disk.
Diffraction sets the ultimate limit on resolution (image sharp-
ness) of optical instruments. According to Rayleigh’s criterion,
two point objects are just barely resolved when their angular
separation u is given by Eq. (36.17). (See Example 36.6.)

Chapter 36 Media Assets

GUIDED PRACTICE For assigned homework and other learning materials, go to Mastering Physics.

KEY EXAMPLE ARIATION PROBLEMS VP36.3.2 You shine a laser on a narrow slit 0.221 mm in width. In
the diffraction pattern that appears on a screen 5.00 m from the slit, the
Be sure to review EXAMPLE 36.1 (Section 36.2) and EXAMPLES 36.2
third minimum is 45.7 mm from the middle of the central bright fringe.
and 36.3 (Section 36.3) before attempting these problems.
Find (a) the wavelength of the laser light and (b) the angle of a line from
VP36.3.1 You pass laser light of wavelength 645 nm through a slit
the center of the slit to the second dark fringe on the screen.
0.250 mm in width and observe the diffraction pattern on a screen a
VP36.3.3 At a certain point in a single-slit diffraction pattern there is a
large distance away. On the screen, the centers of the second minima
phase difference of 35.0 radians between wavelets arriving at the point
on either side of the central bright fringe are 28.0 mm apart. (a) How
from the two edges of the slit. The slit is 0.250 mm wide, and the light
far away is the screen? (b) What would the distance be between these
used has wavelength 545 nm. (a) What is the angle of a line from the cen-
minima if the wavelength were 525 nm?
ter of the slit to this point in the diffraction pattern? (b) If the intensity at
the center of the diffraction pattern is I0, what is the intensity at this point?
 Problems    1209

VP36.3.4 In a single-slit diffraction experiment, the slit width is 0.185 mm would the wavelength have to be to have three strong interference max-
and the wavelength of the light used is 576 nm. (a) Find the angle of ima, with the third at an angle of 88.0°?
a line from the center of the slit to the first dark fringe. (b) You might
Be sure to review EXAMPLE 36.6 (Section 36.7) before attempting
expect the intensity at one-half the angle found in part (a), which is half-
these problems.
way between the middle of the central bright fringe (where the intensity
VP36.6.1 You wish to study the radio emission from the sun at wave-
is I0) and the first dark fringe (where the intensity is zero), to be I0>2.
length 1.70 cm. In order to see details on the sun’s surface no larger than
What is the actual intensity at this angle?
the diameter of the earth, the limit of resolution of the telescope must
Be sure to review EXAMPLE 36.4 (Section 36.5) and EXAMPLE 36.5 be 9.00 * 10-5 rad (about 0.005°). Using the Rayleigh criterion, find
(Section 36.6) before attempting these problems. (a) the minimum diameter your radio telescope must have and (b) the
VP36.5.1 You shine laser light so that it falls normally on a transmis- limit of resolution of this telescope at wavelength 21.1 cm.
sion grating with 825 slits per millimeter. The second-order bright re- VP36.6.2 A telephoto lens for a camera has focal length 0.250 m and
gion occurs at an angle of 41.0° from the central maximum. Find (a) the maximum aperture f>4.0. Take the wavelength of visible light to be
wavelength of the light and (b) the angles for the first-order and third- 5.50 * 10-7 m. (a) Assuming that the resolution is limited by diffrac-
order bright regions. tion, how far away can an object be for you to be able to resolve two
VP36.5.2 The first-order bright region created by a diffraction grating points on the object that are 5.00 mm apart? (b) How far apart are the
occurs at an angle of 24.0° from the central maximum. The wavelength corresponding points on the image made by the lens?
of light that falls normally onto the grating is 625 nm. Find (a) the num- VP36.6.3 A distant point source of light (like a star) emits light of
ber of slits per mm on the grating and (b) the angles of the second-order wavelength 575 nm. When this light enters a camera whose lens has a
bright region and the third-order bright region (if any). focal length of 135 mm, the diffraction pattern formed on the camera’s
VP36.5.3 In a certain crystal the spacing between crystal planes is detector has an Airy disk of radius 0.0112 mm. Find (a) the diameter of
0.165 nm. (a) If you shine a beam of x rays of wavelength 0.124 nm the lens aperture and (b) the f-number.
on this crystal, for what angle between the beam and the crystal planes VP36.6.4 One important goal of astronomers is to have a telescope in
does the first strong interference maximum occur? (b) For what larger space that can resolve planets like the earth orbiting other stars. If a
angles (if any) do strong interference maxima occur? planet orbits its star at a distance of 1.5 * 1011 m (the radius of the
VP36.5.4 In an x-ray diffraction experiment there is only one strong earth’s orbit around the sun) and the telescope has a mirror of diameter
interference maximum, and this occurs when the x-ray beam makes an 8.0 m, how far from the telescope could the star and its planet be if the
angle of 36.0° with the crystal planes. The spacing between the crystal wavelength used was (a) 690 nm and (b) 1400 nm? Use the Rayleigh
planes is 0.158 nm. (a) What is the wavelength of the x rays? (b) What criterion and give your answers in light-years (1 ly = 9.46 * 1015 m).

BRIDGING PROBLEM Observing the Expanding Universe

An astronomer who is studying the light from a galaxy has identified moving source, such as the galaxy, is different from the fre-
the spectrum of hydrogen but finds that the wavelengths are somewhat quency that is emitted. Equation (16.30) relates the emitted fre-
shifted from those found in the laboratory. In the lab, the Ha line in quency, the received frequency, and the velocity of the source
the hydrogen spectrum has a wavelength of 656.3 nm. The astronomer (the target variable). The equation c = fl relates the frequency
is using a transmission diffraction grating having 5758 lines>cm in the ƒ and wavelength l through the speed of light c.
first order and finds that the first bright fringe for the Ha line occurs at
EXECUTE
{23.41° from the central spot. How fast is the galaxy moving? Express
your answer in m>s and as a percentage of the speed of light. Is the gal- 3. Find the wavelength of the Ha spectral line in the received light.
axy moving toward us or away from us? 4. Rewrite Eq. (16.30) as a formula for the velocity v of the galaxy
in terms of the received wavelength and the wavelength emitted
SOLUTION GUIDE by the source.
IDENTIFY and SET UP 5. Solve for v. Express it in m>s and as a percentage of c, and
­decide whether the galaxy is moving toward us or moving away.
1. You can use the information about the grating to find the wave-
length of the Ha line in the galaxy’s spectrum. EVALUATE
2. In Section 16.8 we learned about the Doppler effect for elec- 6. Is your answer consistent with the relative sizes of the received
tromagnetic radiation: The frequency that we receive from a wavelength and the emitted wavelength?

PROBLEMS
•, ••, •••: Difficulty levels. CP: Cumulative problems incorporating material from earlier chapters. CALC: Problems requiring calculus.
DATA: Problems involving real data, scientific evidence, experimental design, and>or statistical reasoning. BIO: Biosciences problems.

DISCUSSION QUESTIONS Q36.3 You use a lens of diameter D and light of wavelength l and
­frequency f to form an image of two closely spaced and distant objects.
Q36.1 Why can we readily observe diffraction effects for sound waves
Which of the following will increase the resolving power? (a) Use a
and water waves, but not for light? Is this because light travels so much
lens with a smaller diameter; (b) use light of higher frequency; (c) use
faster than these other waves? Explain.
light of longer wavelength. In each case justify your answer.
Q36.2 What is the difference between Fresnel and Fraunhofer
­diffraction? Are they different physical processes? Explain.
1210    CHAPTER 36 Diffraction

Q36.4 Light of wavelength l and frequency f passes through a single Q36.18 A hologram is made using 600 nm light and then viewed by
slit of width a. The diffraction pattern is observed on a screen a distance x using white light from an incandescent bulb. What will be seen? Explain.
from the slit. Which of the following will decrease the width of the cen- Q36.19 Ordinary photographic film reverses black and white, in the
tral maximum? (a) Decrease the slit width; (b) decrease the frequency f sense that the most brightly illuminated areas become blackest upon
of the light; (c) decrease the wavelength l of the light; (d) decrease the ­development (hence the term negative). Suppose a hologram nega-
distance x of the screen from the slit. In each case justify your answer. tive is viewed directly, without making a positive transparency. How
Q36.5 In a diffraction experiment with waves of wavelength l, there will the resulting images differ from those obtained with the positive?
will be no intensity minima (that is, no dark fringes) if the slit width is Explain.
small enough. What is the maximum slit width for which this occurs?
Explain your answer.
EXERCISES
Q36.6 An interference pattern is produced by four parallel and equally
spaced narrow slits. By drawing appropriate phasor diagrams, explain Section 36.2 Diffraction from a Single Slit
why there is an interference minimum when the phase difference f 36.1 •• Monochromatic light from a distant source is incident on a slit
from adjacent slits is (a) p>2; (b) p; (c) 3p>2. In each case, for which 0.750 mm wide. On a screen 2.00 m away, the distance from the central
pairs of slits is there totally destructive interference? maximum of the diffraction pattern to the first minimum is measured to
Q36.7 Phasor Diagram for Eight Slits. An interference pattern is pro- be 1.35 mm. Calculate the wavelength of the light.
duced by eight equally spaced narrow slits. The caption for Fig. 36.14 36.2 • Coherent electromagnetic waves with wavelength l pass through a
claims that minima occur for f = 3p>4, p>4, 3p>2, and 7p>4. Draw narrow slit of width a. The diffraction pattern is observed on a tall screen that
the phasor diagram for each of these four cases, and explain why each is 2.00 m from the slit. When l = 500 nm, the width on the screen of the
diagram proves that there is in fact a minimum. In each case, for which central maximum in the diffraction pattern is 8.00 mm. For the same slit and
pairs of slits is there totally destructive interference? screen, what is the width of the central maximum when l = 0.125 mm?
Q36.8 A rainbow ordinarily shows a range of colors (see Section 33.4). 36.3 •• Light of wavelength 585 nm falls on a slit 0.0666 mm wide.
But if the water droplets that form the rainbow are small enough, the (a) On a very large and distant screen, how many totally dark fringes
rainbow will appear white. Explain why, using diffraction ideas. How (indicating complete cancellation) will there be, including both sides of
small do you think the raindrops would have to be for this to occur? the central bright spot? Solve this problem without calculating all the
Q36.9 Some loudspeaker horns for outdoor concerts (at which the angles! (Hint: What is the largest that sin u can be? What does this tell
­entire audience is seated on the ground) are wider vertically than hori- you is the largest that m can be?) (b) At what angle will the dark fringe
zontally. Use diffraction ideas to explain why this is more efficient at that is most distant from the central bright fringe occur?
spreading the sound uniformly over the audience than either a square 36.4 • Light of wavelength 633 nm from a distant source is incident
speaker horn or a horn that is wider horizontally than vertically. Would on a slit 0.750 mm wide, and the resulting diffraction pattern is ob-
this still be the case if the audience were seated at different elevations, served on a screen 3.50 m away. What is the distance between the two
as in an amphitheater? Why or why not? dark fringes on either side of the central bright fringe?
Q36.10 Figure 31.12 (Section 31.2) shows a loudspeaker system. Low- 36.5 •• Diffraction occurs for all types of waves, including sound
frequency sounds are produced by the woofer, which is a speaker with waves. High-frequency sound from a distant source with wavelength
large diameter; the tweeter, a speaker with smaller diameter, produces 9.00 cm passes through a slit 12.0 cm wide. A microphone is placed
high-frequency sounds. Use diffraction ideas to explain why the tweeter 8.00 m directly in front of the center of the slit, corresponding to point O
is more effective for distributing high-frequency sounds uniformly over in Fig. 36.5a. The microphone is then moved in a direction perpendicu-
a room than is the woofer. lar to the line from the center of the slit to point O. At what distances
Q36.11 Information is stored on an audio compact disc, CD-ROM, or from O will the intensity detected by the microphone be zero?
DVD disc in a series of pits on the disc. These pits are scanned by a 36.6 • CP Tsunami! On December 26, 2004, a violent earthquake of
laser beam. An important limitation on the amount of information that magnitude 9.1 occurred off the coast of Sumatra. This quake triggered a
can be stored on such a disc is the width of the laser beam. Explain why huge tsunami (similar to a tidal wave) that killed more than 150,000 people.
this should be, and explain how using a shorter-wavelength laser allows Scientists observing the wave on the open ocean measured the time
more information to be stored on a disc of the same size. between crests to be 1.0 h and the speed of the wave to be 800 km>h.
Q36.12 With which color of light can the Hubble Space Telescope see Computer models of the evolution of this enormous wave showed that
finer detail in a distant astronomical object: red, blue, or ultraviolet? it bent around the continents and spread to all the oceans of the earth.
Explain your answer. When the wave reached the gaps between continents, it diffracted be-
Q36.13 At the end of Section 36.4, the following statements were tween them as through a slit. (a) What was the wavelength of this tsu-
made about an array of N slits. Explain, using phasor diagrams, why nami? (b) The distance between the southern tip of Africa and northern
each statement is true. (a) A minimum occurs whenever f is an integer Antarctica is about 4500 km, while the distance between the southern
multiple of 2p>N (except when f is an integer multiple of 2p, which end of Australia and Antarctica is about 3700 km. As an approximation,
gives a principal maximum). (b) There are 1N - 12 minima between we can model this wave’s behavior by using Fraunhofer diffraction. Find
each pair of principal maxima. the smallest angle away from the central maximum for which the waves
Q36.14 Could x-ray diffraction effects with crystals be observed by would cancel after going through each of these continental gaps.
using visible light instead of x rays? Why or why not? 36.7 •• CP A series of parallel linear water wave fronts are travel-
Q36.15 Why is a diffraction grating better than a two-slit setup for ing directly toward the shore at 15.0 cm>s on an otherwise placid lake.
measuring wavelengths of light? A long concrete barrier that runs parallel to the shore at a distance of
Q36.16 One sometimes sees rows of evenly spaced radio antenna tow- 3.20 m away has a hole in it. You count the wave crests and observe that
ers. A student remarked that these act like diffraction gratings. What did 75.0 of them pass by each minute, and you also observe that no waves
she mean? Why would one want them to act like a diffraction grating? reach the shore at {61.3 cm from the point directly opposite the hole,
Q36.17 If a hologram is made using 600 nm light and then viewed with but waves do reach the shore everywhere within this distance. (a) How
500 nm light, how will the images look compared to those ­observed wide is the hole in the barrier? (b) At what other angles do you find no
when viewed with 600 nm light? Explain. waves hitting the shore?
 Problems    1211

36.8 • Monochromatic electromagnetic radiation with wavelength l 36.18 • Parallel rays of monochromatic light with wavelength 568 nm il-
from a distant source passes through a slit. The diffraction pattern is ob- luminate two identical slits and produce an interference pattern on a screen
served on a screen 2.50 m from the slit. If the width of the central maxi- that is 75.0 cm from the slits. The centers of the slits are 0.640 mm apart
mum is 6.00 mm, what is the slit width a if the wavelength is (a) 500 nm and the width of each slit is 0.434 mm. If the intensity at the center of the
(visible light); (b) 50.0 mm (infrared radiation); (c) 0.500 nm (x rays)? central maximum is 5.00 * 10-4 W>m2, what is the intensity at a point
36.9 •• Doorway Diffraction. Sound of frequency 1250 Hz leaves a on the screen that is 0.900 mm from the center of the central maximum?
room through a 1.00-m-wide doorway (see Exercise 36.5). At which angles 36.19 • Number of Fringes in a Diffraction Maximum. In
relative to the centerline perpendicular to the doorway will someone out- Fig. 36.12c the central diffraction maximum contains exactly seven in-
side the room hear no sound? Use 344 m>s for the speed of sound in air and terference fringes, and in this case d>a = 4. (a) What must the ratio d>a
assume that the source and listener are both far enough from the doorway be if the central maximum contains exactly five fringes? (b) In the case
for Fraunhofer diffraction to apply. You can ignore effects of reflections. considered in part (a), how many fringes are contained within the first
36.10 • CP Light waves, for which the electric field is given by diffraction maximum on one side of the central maximum?
Ey1x, t2 = Emax sin311.40 * 107 m-12x - vt4, pass through a slit and 36.20 •• Diffraction and Interference Combined. Consider the
produce the first dark bands at {28.6° from the center of the diffraction ­interference pattern produced by two parallel slits of width a and ­separation
pattern. (a) What is the frequency of this light? (b) How wide is the slit? d, in which d = 3a. The slits are illuminated by normally incident light of
(c) At which angles will other dark bands occur? wavelength l. (a) First we ignore diffraction effects due to the slit width. At
36.11 •• Red light of wavelength 633 nm from a helium–neon laser what angles u from the central maximum will the next four maxima in the
passes through a slit 0.350 mm wide. The diffraction pattern is observed two-slit interference pattern occur? Your answer will be in terms of d and l.
on a screen 3.00 m away. Define the width of a bright fringe as the dis- (b) Now we include the effects of diffraction. If the intensity at u = 0° is I0,
tance between the minima on either side. (a) What is the width of the what is the intensity at each of the angles in part (a)? (c) Which double-slit
central bright fringe? (b) What is the width of the first bright fringe on interference maxima are missing in the pattern? (d) Compare your results
either side of the central one? to those illustrated in Fig. 36.12c. In what ways are your results different?
36.12 • When coherent electromagnetic waves with wavelength 36.21 •• An interference pattern is produced by light of wavelength
l = 120 mm are incident on a single slit of width a, the width of the cen- 580 nm from a distant source incident on two identical parallel slits sep-
tral maximum on a tall screen 1.50 m from the slit is 90.0 cm. For the same arated by a distance (between centers) of 0.530 mm. (a) If the slits are
slit and screen, for what wavelength of the incident waves is the width of very narrow, what would be the angular positions of the first-order and
the central maximum 180.0 cm, double the value when l = 120 mm? second-order, two-slit interference maxima? (b) Let the slits have width
0.320 mm. In terms of the intensity I0 at the center of the central maxi-
Section 36.3 Intensity in the Single-Slit Pattern mum, what is the intensity at each of the angular positions in part (a)?
36.13 •• Monochromatic light of wavelength 580 nm passes through 36.22 •• Laser light of wavelength 500.0 nm illuminates two identi-
a single slit and the diffraction pattern is observed on a screen. Both the cal slits, producing an interference pattern on a screen 90.0 cm from the
source and screen are far enough from the slit for Fraunhofer diffraction slits. The bright bands are 1.00 cm apart, and the third bright bands on
to apply. (a) If the first diffraction minima are at {90.0°, so the cen- either side of the central maximum are missing in the pattern. Find the
tral maximum completely fills the screen, what is the width of the slit? width and the separation of the two slits.
(b) For the width of the slit as calculated in part (a), what is the ratio of 36.23 • Coherent electromagnetic waves with wavelength
the intensity at u = 45.0° to the intensity at u = 0? l = 500 nm pass through two identical slits. The width of each slit is
36.14 •• Monochromatic light of wavelength l = 620 nm from a dis- a, and the distance between the centers of the slits is d = 9.00 mm.
tant source passes through a slit 0.450 mm wide. The diffraction pattern (a) What is the smallest possible width a of the slits if the m = 3 max-
is observed on a screen 3.00 m from the slit. In terms of the intensity I0 imum in the interference pattern is not present? (b) What is the next
at the peak of the central maximum, what is the intensity of the light at larger value of the slit width for which the m = 3 maximum is absent?
the screen the following distances from the center of the central maxi- 36.24 •• Coherent light with wavelength 200 nm passes through two
mum: (a) 1.00 mm; (b) 3.00 mm; (c) 5.00 mm? identical slits. The width of each slit is a, and the distance between the
36.15 •• Public Radio station KXPR-FM in Sacramento broadcasts at centers of the slits is d = 1.00 mm. The m = 5 maximum in the two-
88.9 MHz. The radio waves pass between two tall skyscrapers that are slit interference pattern is absent, but the maxima for m = 0 through
15.0 m apart along their closest walls. (a) At what horizontal angles, m = 4 are present. What is the ratio of the intensities for the m = 1 and
relative to the original direction of the waves, will a distant antenna m = 2 maxima in the two-slit pattern?
not receive any signal from this station? (b) If the maximum intensity
is 3.50 W>m2 at the antenna, what is the intensity at {5.00° from the Section 36.5 The Diffraction Grating
­center of the central maximum at the distant antenna? 36.25 • When laser light of wavelength 632.8 nm passes through a
36.16 • Monochromatic light of wavelength 592 nm from a distant diffraction grating, the first bright spots occur at {17.8° from the cen-
source passes through a slit that is 0.0290 mm wide. In the resulting tral maximum. (a) What is the line density (in lines>cm) of this grating?
diffraction pattern, the intensity at the center of the central maximum (b) How many additional bright spots are there beyond the first bright
1u = 0°2 is 4.00 * 10-5 W>m2. What is the intensity at a point on the spots, and at what angles do they occur?
screen that corresponds to u = 1.20°? 36.26 •• Monochromatic light is at normal incidence on a plane trans-
mission grating. The first-order maximum in the interference pattern is at an
Section 36.4 Multiple Slits angle of 11.3°. What is the angular position of the fourth-order maximum?
36.17 • Nearly monochromatic coherent light waves leave two rectan- 36.27 • You send coherent 550 nm light through a diffraction grating
gular slits in phase and at an angle of u = 22.0° with the normal. When that has slits of equal widths and constant separation between adjacent
the light reaches a distant screen, the waves from the center of one slit slits. You expect to see the fourth-order interference maximum at an
are 344° out of phase with the waves from the center of the other slit, and angle of 66.6° with respect to the normal to the grating. However, that
the waves from the top of either slit are 172° out of phase with the waves order is missing because 66.6° is also the angle for the third diffrac-
from the bottom of that slit. (a) How is the center-to-center distance be- tion minimum (as measured from the central diffraction maximum) for
tween the slits related to the width of either slit? (b) Calculate the inten- each slit. (a) Find the center-to-center distance between adjacent slits.
sity at the screen for u = 22.0° if the intensity at u = 0° is 0.234 W>m2. (b) Find the number of slits per mm. (c) Find the width of each slit.
1212    CHAPTER 36 Diffraction

36.28 • If a diffraction grating produces a third-order bright spot for red light 36.40 • BIO If you can read the bottom row of your doctor’s 1
eye
(of wavelength 700 nm) at 65.0° from the central maximum, at what angle chart, your eye has a resolving power of 1 arcminute, equal to 60 degree.
will the second-order bright spot be for violet light (of wavelength 400 nm)? If this resolving power is diffraction limited, to what effective diameter
36.29 • Two Fraunhofer lines in the solar absorption spectrum have of your eye’s optical system does this correspond? Use Rayleigh’s crite-
wavelengths of 430.790 nm and 430.774 nm. A diffraction grating rion and assume l = 550 nm.
has 12,800 slits. (a) What is the minimum chromatic resolving power 36.41 •• The VLBA (Very Long Baseline Array) uses a number of in-
needed to resolve these two spectral lines? (b) What is the lowest order dividual radio telescopes to make one unit having an equivalent diameter
required to resolve these two lines? of about 8000 km. When this radio telescope is focusing radio waves of
36.30 • The wavelength range of the visible spectrum is ­approximately wavelength 2.0 cm, what would have to be the diameter of the mirror of
380–750 nm. White light falls at normal incidence on a diffraction ­grating a visible-light telescope focusing light of wavelength 550 nm so that the
that has 350 slits>mm. Find the angular width of the visible spectrum in visible-light telescope has the same resolution as the radio telescope?
(a) the first order and (b) the third order. (Note: An advantage of working 36.42 • Photography. A wildlife photographer uses a moderate
in higher orders is the greater angular spread and ­better resolution. A disad- telephoto lens of focal length 135 mm and maximum aperture f>4.00
vantage is the overlapping of different orders, as shown in Example 36.4.) to photograph a bear that is 11.5 m away. Assume the wavelength is
36.31 • (a) What is the wavelength of light that is deviated in the 550 nm. (a) What is the width of the smallest feature on the bear that
first order through an angle of 13.5° by a transmission grating having this lens can resolve if it is opened to its maximum aperture? (b) If,
5000 slits>cm? (b) What is the second-order deviation of this wave- to gain depth of field, the photographer stops the lens down to f>22.0,
length? Assume normal incidence. what would be the width of the smallest resolvable feature on the bear?
36.32 •• CDs and DVDs as Diffraction Gratings. A laser beam of 36.43 •• Hubble Versus Arecibo. The Hubble Space Telescope has an
wavelength l = 632.8 nm shines at normal incidence on the reflective aperture of 2.4 m and focuses visible light (380–750 nm). The Arecibo
side of a compact disc. (a) The tracks of tiny pits in which information radio telescope in Puerto Rico is 305 m (1000 ft) in diameter (it is built in a
is coded onto the CD are 1.60 mm apart. For what angles of reflection mountain valley) and focuses radio waves of wavelength 75 cm. (a) Under
(measured from the normal) will the intensity of light be maximum? optimal viewing conditions, what is the smallest crater that each of these
(b) On a DVD, the tracks are only 0.740 mm apart. Repeat the calcula- telescopes could resolve on our moon? (b) If the Hubble Space Telescope
tion of part (a) for the DVD. were to be converted to surveillance use, what is the highest orbit above
36.33 • Identifying Isotopes by Spectra. Different isotopes of the the surface of the earth it could have and still be able to resolve the license
same element emit light at slightly different wavelengths. A wavelength plate (not the letters, just the plate) of a car on the ground? Assume opti-
in the emission spectrum of a hydrogen atom is 656.45 nm; for deute- mal viewing conditions, so that the resolution is diffraction limited.
rium, the corresponding wavelength is 656.27 nm. (a) What minimum 36.44 • Observing Jupiter. You are asked to design a space telescope
number of slits is required to resolve these two wavelengths in second for earth orbit. When Jupiter is 5.93 * 108 km away (its closest ap-
order? (b) If the grating has 500.00 slits>mm, find the angles and angu- proach to the earth), the telescope is to resolve, by Rayleigh’s criterion,
lar separation of these two wavelengths in the second order. features on Jupiter that are 250 km apart. What minimum-diameter mir-
ror is required? Assume a wavelength of 500 nm.
Section 36.6 X-Ray Diffraction
36.34 • If the planes of a crystal are 3.50 Å (1 Å = 10-10 m = 1 PROBLEMS
Ångstrom unit) apart, (a) what wavelength of electromagnetic waves is
needed so that the first strong interference maximum in the Bragg reflec- 36.45 •• BIO Thickness of Human Hair. Although we have dis-
tion occurs when the waves strike the planes at an angle of 22.0°, and cussed single-slit diffraction only for a slit, a similar result holds when
in what part of the electromagnetic spectrum do these waves lie? (See light bends around a straight, thin object, such as a strand of hair. In that
Fig. 32.4.) (b) At what other angles will strong interference maxima occur? case, a is the width of the strand. From actual laboratory measurements
36.35 • X rays of wavelength 0.0850 nm are scattered from the atoms on a human hair, it was found that when a beam of light of wavelength
of a crystal. The second-order maximum in the Bragg reflection occurs 632.8 nm was shone on a single strand of hair, and the diffracted light
when the angle u in Fig. 36.22 is 21.5°. What is the spacing between was viewed on a screen 1.25 m away, the first dark fringes on either
adjacent atomic planes in the crystal? side of the central bright spot were 5.22 cm apart. How thick was this
36.36 • Monochromatic x rays are incident on a crystal for which the strand of hair?
spacing of the atomic planes is 0.440 nm. The first-order maximum in 36.46 •• CP A loudspeaker with a diaphragm that vibrates at 960 Hz
the Bragg reflection occurs when the incident and reflected x rays make is traveling at 80.0 m>s directly toward a pair of holes in a very large
an angle of 39.4° with the crystal planes. What is the wavelength of the wall. The speed of sound in the region is 344 m>s. Far from the wall,
x rays? you observe that the sound coming through the openings first cancels
at {11.4° with respect to the direction in which the speaker is moving.
Section 36.7 Circular Apertures and Resolving Power (a) How far apart are the two openings? (b) At what angles would the
36.37 •• Monochromatic light with wavelength 620 nm passes sound first cancel if the source stopped moving?
through a circular aperture with diameter 7.4 mm. The resulting diffrac- 36.47 ••• Laser light of wavelength 632.8 nm falls normally
tion pattern is observed on a screen that is 4.5 m from the aperture. on a slit that is 0.0250 mm wide. The transmitted light is viewed
What is the diameter of the Airy disk on the screen? on a distant screen where the intensity at the center of the central
36.38 •• Monochromatic light with wavelength 490 nm passes through bright fringe is 8.50 W>m2. (a) Find the maximum number of totally
a circular aperture, and a diffraction pattern is observed on a screen that dark fringes on the screen, assuming the screen is large enough to
is 1.20 m from the aperture. If the distance on the screen between the first show them all. (b) At what angle does the dark fringe that is most
and second dark rings is 1.65 mm, what is the diameter of the aperture? ­distant from the center occur? (c) What is the maximum intensity
36.39 • Two satellites at an altitude of 1200 km are separated of the bright fringe that occurs immediately before the dark fringe
by 28 km. If they broadcast 3.6 cm microwaves, what minimum in part (b)? Approximate the angle at which this fringe occurs by
­receiving-dish diameter is needed to resolve (by Rayleigh’s criterion) ­assuming it is midway between the angles to the dark fringes on
the two transmissions? ­either side of it.
 Problems    1213

36.48 • Grating Design. Your boss asks you to design a diffraction 36.56 •• Humans perceive sound with frequencies in the range 20
grating that will disperse the first-order visible spectrum through an an- to 20,000 Hz. Speech lies in the middle of this range, 400 to 1000 Hz,
gular range of 27.0°. (See Example 36.4 in Section 36.5.) (a) What must while bells, sirens, and the knock of silverware on plates all extend
be the number of slits per centimeter for this grating? (b) At what angles above the range for speech. Musical instruments that keep rhythm,
will the first-order visible spectrum begin and end? such as drums and bass guitars, supply sounds with lower frequen-
36.49 • Measuring Refractive Index. A thin slit illuminated by light cies. Doorways act as natural filters, so noises sound different when
of frequency f produces its first dark band at {38.2° in air. When the they come from around corners. (a) Estimate the width of a typical
entire apparatus (slit, screen, and space in between) is immersed in an doorway in your house. (b) Determine the sound frequency that corre-
unknown transparent liquid, the slit’s first dark bands occur instead at sponds to a wavelength comparable to the width of a doorway. (Such
{21.6°. Find the refractive index of the liquid. waves freely diffract around the corner of the door.) (c) Determine the
36.50 •• Underwater Photography. An underwater camera has a frequency of sound that would diffract only {20° from directly for-
lens with focal length in air of 35.0 mm and a maximum aperture of ward when passing through a doorway. (d) Based on your estimates,
f>2.80. The film it uses has an emulsion that is sensitive to light of fre- does a doorway act as a noticeable filter for sounds we can hear?
quency 6.00 * 1014 Hz. If the photographer takes a picture of an object (e) If so, which frequency range (low, middle, high) is harder to hear
2.75 m in front of the camera with the lens wide open, what is the width from around the corner of a doorway? (f) Experiment to see if you can
of the smallest resolvable detail on the subject if the object is (a) a fish notice this effect. It is more readily perceived from outside a house as
underwater with the camera in the water and (b) a person on the beach you listen to sounds from inside. Why is this so?
with the camera out of the water? 36.57 • A diffraction grating has 650 slits>mm. What is the highest
36.51 ••• CALC The intensity of light in the Fraunhofer diffraction order that contains the entire visible spectrum? (The wavelength range
pattern of a single slit is given by Eq. (36.5). Let g = b>2. (a) Show of the visible spectrum is approximately 380–750 nm.)
that the equation for the values of g at which I is a maximum is 36.58 •• Quasars, an abbreviation for quasi-stellar radio sources,
tan g = g. (b) Determine the two smallest positive values of g that are distant objects that look like stars through a telescope but that emit
are solutions of this equation. (Hint: You can use a trial-and-error far more electromagnetic radiation than an entire normal galaxy of
procedure. Guess a value of g and adjust your guess to bring tan g stars. An example is the bright object below and to the left of center in
closer to g. A graphical solution of the equation is very helpful in Fig. P36.58; the other elongated objects in this image are normal galax-
locating the solutions approximately, to get good initial guesses.) ies. The leading model for the structure of a quasar is a galaxy with a su-
(c) What are the positive values of g for the first, second, and third permassive black hole at its center. In this model, the radiation is emitted
minima on one side of the central maximum? Are the g values in by interstellar gas and dust within the galaxy as this material falls toward
part (b) precisely halfway between the g values for adjacent minima? the black hole. The radiation is thought to emanate from a region just
(d) If a = 12l, what are the angles u (in degrees) that locate the first a few light-years in diameter. (The diffuse glow surrounding the bright
minimum, the first maximum beyond the central maximum, and the quasar shown in Fig. P36.58 is thought to be this quasar’s host galaxy.)
second minimum? To investigate this model of quasars and to study other exotic astronomi-
36.52 •• A slit 0.360 mm wide is illuminated by parallel rays of light cal objects, the Russian Space Agency has placed a radio telescope in
that have a wavelength of 540 nm. The diffraction pattern is observed a large orbit around the earth. When this telescope is 77,000 km from
on a screen that is 1.20 m from the slit. The intensity at the center of the earth and the signals it receives are combined with signals from the
central maximum 1u = 0°2 is I0. (a) What is the distance on the screen ground-based telescopes of the VLBA, the resolution is that of a single
from the center of the central maximum to the first minimum? (b) What radio telescope 77,000 km in diameter. What is the size of the small-
is the distance on the screen from the center of the central maximum to est detail that this arrangement can resolve in quasar 3C 405, which is
the point where the intensity has fallen to I0>2? 7.2 * 108 light-years from earth, using radio waves at a frequency of
36.53 •• CP CALC In a large vacuum chamber, monochromatic laser 1665 MHz? (Hint: Use Rayleigh’s criterion.) Give your answer in light-
light passes through a narrow slit in a thin aluminum plate and forms years and in kilometers.
a diffraction pattern on a screen that is 0.620 m from the slit. When the
aluminum plate has a temperature of 20.0°C, the width of the central
maximum in the diffraction pattern is 2.75 mm. What is the change in Figure P36.58
the width of the central maximum when the temperature of the plate is
raised to 520.0°C? Does the width of the central diffraction ­maximum
increase or decrease when the temperature is increased?
36.54 •• CP In a laboratory, light from a particular spectrum line of
helium passes through a diffraction grating and the second-order maxi-
mum is at 18.9° from the center of the central bright fringe. The same
grating is then used for light from a distant galaxy that is moving away
from the earth with a speed of 2.65 * 107 m>s. For the light from the
galaxy, what is the angular location of the second-order maximum for
the same spectral line as was observed in the lab? (See Section 16.8.)
36.55 • What is the longest wavelength that can be observed in the
third order for a transmission grating having 9200 slits>cm? Assume
normal incidence.
1214    CHAPTER 36 Diffraction

36.59 •• In the 1920s Clinton Davisson and Lester Germer acciden- 36.62 •• DATA Your physics study partner tells you that the width of
tally observed diffraction when electrons with 54 eV of energy were the central bright band in a single-slit diffraction pattern is inversely
scattered off crystalline nickel. The diffraction peak occurred when proportional to the width of the slit. This means that the width of the
the angle between the incident beam and the scattered beam was 50°. central maximum increases when the width of the slit decreases. The
(a) What is the corresponding angle u relevant for Eq. (36.16)? (b) The claim seems counterintuitive to you, so you make measurements to test
planes in crystalline nickel are separated by 0.091 nm, as determined it. You shine monochromatic laser light with wavelength l onto a very
by x-ray scattering experiments. According to the Bragg condition, narrow slit of width a and measure the width w of the central maximum
what wavelength do the electrons in these experiments have? (c) Given in the diffraction pattern that is produced on a screen 1.50 m from the
the mass of an electron as 9.11 * 10-31 kg, what is the corresponding slit. (By “width,” you mean the distance on the screen between the two
classical speed vcl of the diffracted electrons? (d) Assuming the elec- minima on either side of the central maximum.) Your measurements are
trons correspond to a wave with speed vcl and wavelength l, what is given in the table.
the frequency f of the diffracted waves? (e) Quantum mechanics pos-
tulates that the energy E and the frequency f of a particle are related by
E = hf , where h is known as Planck’s constant. Estimate h from these a (Mm) 0.78 0.91 1.04 1.82 3.12 5.20 7.80 10.40 15.60
observations. (f) Our analysis has a small flaw: The relevant wave ve- w (m) 2.68 2.09 1.73 0.89 0.51 0.30 0.20 0.15 0.10
locity, known as a quantum phase velocity, is half the classical particle
velocity, for reasons explained by deeper aspects of quantum physics.
Re-estimate the value of h using this modification. (g) The established (a) If w is inversely proportional to a, then the product aw is constant,
value of Planck’s constant is 6.6 * 10-34 J # s. Does this agree with independent of a. For the data in the table, graph aw versus a. Explain
your estimate? why aw is not constant for smaller values of a. (b) Use your graph in
36.60 •• BIO Resolution of the Eye. The maximum resolution of part (a) to calculate the wavelength l of the laser light. (c) What is
the eye depends on the diameter of the opening of the pupil (a diffrac- the angular position of the first minimum in the diffraction pattern for
tion ­effect) and the size of the retinal cells. The size of the retinal cells (i) a = 0.78 mm and (ii) a = 15.60 mm?
(about 5.0 mm in diameter) limits the size of an object at the near point 36.63 •• DATA At the metal fabrication company where you work,
(25 cm) of the eye to a height of about 50 mm. (To get a reasonable you are asked to measure the diameter D of a very small circular hole
­estimate without having to go through complicated calculations, we in a thin, vertical metal plate. To do so, you pass coherent monochro-
shall ignore the effect of the fluid in the eye.) (a) Given that the diam- matic light with wavelength 562 nm through the hole and observe the
eter of the human pupil is about 2.0 mm, does the Rayleigh criterion diffraction pattern on a screen that is a distance x from the hole. You
allow us to resolve a 50@mm@tall object at 25 cm from the eye with light measure the radius r of the first dark ring in the diffraction pattern (see
of wavelength 550 nm? (b) According to the Rayleigh criterion, what is Fig. 36.26). You make the measurements for four values of x. Your
the shortest object we could resolve at the 25 cm near point with light ­results are given in the table.
of wavelength 550 nm? (c) What angle would the object in part (b) sub-
tend at the eye? Express your answer in minutes 160 min = 1°2, and
compare it with the experimental value of about 1 min. (d) Which effect x (m) 1.00 1.50 2.00 2.50
is more important in limiting the resolution of our eyes: diffraction or r (cm) 5.6 8.5 11.6 14.1
the size of the retinal cells?
36.61 •• DATA While researching the use of laser pointers, you
conduct a diffraction experiment with two thin parallel slits. Your re- (a) Use each set of measurements to calculate D. Because the mea-
sult is the pattern of closely spaced bright and dark fringes shown in surements contain some error, calculate the average of the four values
Fig. P36.61. (Only the central portion of the pattern is shown.) You of D and take that to be your reported result. (b) For x = 1.00 m,
measure that the bright spots are equally spaced at 1.53 mm center to what are the radii of the second and third dark rings in the diffraction
center (except for the missing spots) on a screen that is 2.50 m from pattern?
the slits. The light source was a helium–neon laser producing a wave- 36.64 •• A glass sheet is covered by a very thin opaque coating. In
length of 632.8 nm. (a) How far apart are the two slits? (b) How wide the middle of this sheet there is a thin scratch 0.00125 mm thick. The
is each one? sheet is totally immersed beneath the surface of a liquid. Parallel rays of
Figure P36.61 monochromatic coherent light with wavelength 612 nm in air strike the
sheet perpendicular to its surface and pass through the scratch. A screen
1.53 mm is placed in the liquid a distance of 30.0 cm away from the sheet and
parallel to it. You observe that the first dark fringes on either side of the
central bright fringe on the screen are 22.4 cm apart. What is the refrac-
tive index of the liquid?
 Problems    1215

CHALLENGE PROBLEMS Then, using the relationship eiz = cos z + i sin z, show that the (real)
36.65 ••• An opaque barrier has Figure P36.65 electric field at point P is
an inner membrane and an outer sin1Nf>22
membrane that slide past each other, EP1t2 = c E0 d cos 3kR - vt + 1N - 12f>24
P sin1f>22
as shown in Fig. P36.65. Each
membrane includes parallel slits of The quantity in the first square brackets in this expression is the am-
width a separated by a distance d. d plitude of the electric field at P. (d) Use the result for the electric-field
A screen forms a circular arc sub- Q amplitude in part (c) to show that the intensity at an angle u is
u
tending 60° at the fixed midpoint
sin1Nf>22 2
between the slits. A green 532 nm I = I0 c d
laser impinges on the slits from the a S sin1f>22
left. The outer membrane moves v S
v
where I0 is the maximum intensity for an individual slit. (e) Check
upward with speed v while the the result in part (d) for the case N = 2. It will help to recall that
inner membrane moves downward Not to scale sin 2A = 2 sin A cos A. Explain why your result differs from Eq. (35.10),
with the same speed, propelled by the expression for the intensity in two-source interference, by a factor of 4.
nanomotors. At time t = 0, point P (Hint: Is I0 defined in the same way in both expressions?)
on the outer membrane is adjacent to point Q on the inner membrane so 36.67 ••• CALC Intensity Pattern of N Slits, Continued. Part (d)
that the effective aperture width is zero. The aperture is fully closed again of Challenge Problem 36.66 gives an expression for the intensity in the
at t = 3.00 s. (a) At t = 1.00 s, there are 19 evenly spaced bright spots on interference pattern of N identical slits. Use this result to verify the fol-
the screen, each of approximately the same intensity. At the edges of the lowing statements. (a) The maximum intensity in the pattern is N 2I0.
screen the first diffraction minimum and a two-slit interference maximum (b) The principal maximum at the center of the pattern extends from
coincide. What is the slit distance d? (Note: The screen does not encom- f = -2p>N to f = 2p>N, so its width is inversely proportional to
pass the entire diffraction pattern.) (b) What is the speed v? (c) What is 1>N. (c) A minimum occurs whenever f is an integer multiple of 2p>N,
the maximum aperture width a? (d) At a certain time, the outermost spots except when f is an integer multiple of 2p (which gives a principal
(the m = {9 spots) disappear. What is that time? (e) At t = 1.50 s, maximum). (d) There are 1N - 12 minima between each pair of princi-
what is the intensity of the m = {1 spots in terms of the m = 0 central pal maxima. (e) Halfway between two principal maxima, the intensity
spot? (f) What are the angular positions of these spots? can be no greater than I0; that is, it can be no greater than 1>N 2 times the
36.66 ••• Intensity Pattern of N Slits. (a) Consider an arrangement intensity at a principal maximum.
of N slits with a distance d between adjacent slits. The slits emit coher- 36.68 ••• CALC It is possible to calculate the intensity in the single-
ently and in phase at wavelength l. Show that at a time t, the electric slit Fraunhofer diffraction pattern without using the phasor method of
field at a distant point P is Section 36.3. Let y′ represent the position of a point within the slit of
EP1t2 = E0 cos1kR - vt2 + E0 cos1kR - vt + f2 width a in Fig. 36.5a, with y′ = 0 at the center of the slit so that the slit
extends from y′ = -a>2 to y′ = a>2. We imagine dividing the slit up
+ E0 cos1kR - vt + 2f2 + g
into infinitesimal strips of width dy′, each of which acts as a source of
+ E0 cos1kR - vt + 1N - 12f2 secondary wavelets. (a) The amplitude of the total wave at the point O
on the distant screen in Fig. 36.5a is E0 . Explain why the amplitude of
where E0 is the amplitude at P of the electric field due to an indi-
the wavelet from each infinitesimal strip within the slit is E01dy′>a2,
vidual slit, f = 12pd sin u2>l, u is the angle of the rays reaching P
so that the electric field of the wavelet a distance x from the infinitesi-
(as measured from the perpendicular bisector of the slit arrangement),
mal strip is dE = E01dy′>a2 sin 1kx - vt2. (b) Explain why the wavelet
and R is the distance from P to the most distant slit. In this prob-
from each strip as detected at point P in Fig. 36.5a can be expressed as
lem, assume that R is much larger than d. (b) To carry out the sum
in part (a), it is convenient to use the complex-number relationship dy′
eiz = cos z + i sin z, where i = 1-1. In this expression, cos z is the dE = E0 sin3k1D - y′ sin u2 - vt4
a
real part of the complex number eiz, and sin z is its imaginary part.
Show that the electric field EP1t2 is equal to the real part of the com- where D is the distance from the center of the slit to point P and
plex quantity k = 2p>l. (c) By integrating the contributions dE from all parts of the
slit, show that the total wave detected at point P is

a E0 e
N-1
i1kR - vt + nf2
sin3ka1sin u2>24
n=0 E = E0 sin1kD - vt2
ka1sin u2>2
(c) Using the properties of the exponential function that eAeB = e1A + B2
sin3pa1sin u2>l4
and 1eA2 n = enA, show that the sum in part (b) can be written as = E0 sin1kD - vt2
pa1sin u2>l
(The trigonometric identities in Appendix B will be useful.) Show
eiNf - 1 i1kR - vt2 that at u = 0°, corresponding to point O in Fig. 36.5a, the wave is
E0 a be
eif - 1 E = E0 sin1kD - vt2 and has amplitude E0 , as stated in part (a).
(d) Use the result of part (c) to show that if the intensity at point O is I0 ,
eiNf>2 - e-iNf>2
= E0 a bei3kR - vt + 1N - 12f>24 then the intensity at a point P is given by Eq. (36.7).
eif>2 - e-if>2
1216    CHAPTER 36 Diffraction

MCAT-STYLE PASSAGE PROBLEMS spacing of the microspheres allows the longer-wavelength visible light
to produce more destructive interference than can x rays. (c) The micro-
Bragg Reflection on a Different Scale. A colloid consists of particles
spheres are much larger than atoms in a crystalline solid, and in order to
of one type of substance dispersed in another substance. Suspensions of
get interference maxima at reasonably large angles, the wavelength must
electrically charged microspheres (microscopic spheres, such as poly-
be much longer than the size of the individual scatterers. (d) The micro-
styrene) in a liquid such as water can form a colloidal crystal when the
spheres are spaced more widely than atoms in a crystalline solid, and in
microspheres arrange themselves in a regular repeating pattern under
order to get interference maxima at reasonably large angles, the wave-
the influence of the electrostatic force. Colloidal crystals can selec-
length must be comparable to the spacing between scattering planes.
tively manipulate different wavelengths of visible light. Just as we can
36.70 What plane spacing in the colloidal crystal could produce the
study crystalline solids by using Bragg reflection of x rays, we can
maximum in this experiment? (a) 390 nm; (b) 520 nm; (c) 650 nm;
study colloidal crystals through Bragg scattering of visible light from
(d) 780 nm.
the regular arrangement of charged microspheres. Because the light is
36.71 When the light is passed through the bottom of the sample con-
traveling through a liquid when it experiences the path differences that
tainer, the interference maximum is observed to be at 41°; when it is
lead to constructive interference, it is the wavelength in the liquid that
passed through the top, the corresponding maximum is at 37°. What
determines the angles at which Bragg reflections are seen. In one ex-
is the best explanation for this observation? (a) The microspheres are
periment, laser light with a wavelength in vacuum of 650 nm is passed
more tightly packed at the bottom, because they tend to settle in the
through a sample of charged polystyrene spheres in water. A strong in-
suspension. (b) The microspheres are more tightly packed at the top,
terference maximum is then observed when the incident and reflected
because they tend to float to the top of the suspension. (c) The increased
beams make an angle of 39° with the colloidal crystal planes.
pressure at the bottom makes the microspheres smaller there. (d) The
36.69 Why is visible light, which has much longer wavelengths than
maximum at the bottom corresponds to m = 2, whereas the maximum
x rays do, used for Bragg reflection experiments on colloidal crystals?
at the top corresponds to m = 1.
(a) The microspheres are suspended in a liquid, and it is more difficult
for x rays to penetrate liquid than it is for visible light. (b) The irregular

ANSWERS
Chapter Opening Question
?
(i) For an optical system that uses a lens, the ability to resolve fine
VP36.5.1 (a) 398 nm (a) 19.1° for m = 1, 79.8° for m = 3
VP36.5.2 (a) 651 slits>mm (b) 54.4° for m = 2; there is no m = 3
bright region
details—its resolving power, or resolution—improves as the lens di-
VP36.5.3 (a) 22.1° (b) 48.7°
ameter D increases (Section 36.7). Each miniature lens in a fly’s eye
VP36.5.4 (a) 0.186 nm (b) 0.105 nm
produces its own image, so these images have very poor resolution,
VP36.6.1 (a) 2.50 * 102 m (b) 1.11 * 10-3 rad = 0.0640°
compared to those produced by a human eye, because the lens is so
VP36.6.2 (a) 466 m (b) 0.00268 mm
small. However, a fly’s eye is much better than a human eye at detect-
VP36.6.3 (a) 8.46 mm (b) f>16.0
ing movement.
VP36.6.4 (a) 150 ly (b) 74 ly
Key Example ARIATION Problems
Bridging Problem
VP36.3.1 (a) 2.30 m (b) 22.8 mm
1.501 * 107 m>s or 5.00% of c; away from us
VP36.3.2 (a) 673 nm (b) 0.349°
VP36.3.3 (a) 0.696° (b) 13.11 * 10-32I0
VP36.3.4 (a) 0.178° (b) 0.405I0