The beautiful colors from the surface of this soap bubble can be nicely explained by the wave theory of light. A soap bubble is a very thin spherical film filled with air. Light reflected from the outer and inner surfaces of this thin film of soapy water interferes constructively to produce the bright colors. Which color we see at any point depends on the thickness of the soapy water film at that point and also on the viewing angle. Near the top of the bubble, we see a small black area surrounded by a silver or white area. The bubble’s thickness is smallest at that black spot, perhaps only about 30 nm thick, and is fully transparent (we see the black background).

We cover fundamental aspects of the wave nature of light, including two-slit interference and interference in thin films.** T ** 34

CONTENTS34-1 Waves versus Particles;

Huygens’ Principle and Diffraction

34-2 Huygens’ Principle and the Law of Refraction

34-3 Interference—Young’s Double-Slit Experiment

*34-4 Intensity in the Double-Slit Interference Pattern

34-5 Interference in Thin Films*34-6 Michelson Interferometer*34-7 Luminous Intensity

The Wave Nature of Light; InterferenceCHAPTER-OPENING QUESTIOI — Guess now!When a thin layer of o il lies on top of water or wet pavement, you can often see swirls of color. We also see swirls of color on the soap bubble shown above. What causes these colors?

(a) Additives in the o il or soap reflect various colors.(b ) Chemicals in the o il or soap absorb various colors.(c) Dispersion due to differences in index o f refraction in the o il or soap.(d ) The interactions of the light w ith a thin boundary layer where the o il (or

soap) and the water have mixed irregularly.(e) Light waves reflected from the top and bottom surfaces of the thin o il or

soap film can add up constructively fo r particular wavelengths.

That light carries energy is obvious to anyone who has focused the Sun’s rays w ith a magnifying glass on a piece of paper and burned a hole in it. But how does light travel, and in what form is this energy carried? In our discussion of waves in Chapter 15, we noted that energy can be carried

from place to place in basically two ways: by particles or by waves. In the first case, material objects or particles can carry energy, such as an avalanche of rocks or rushing water. In the second case, water waves and sound waves, for example, can carry energy over long distances even though the oscillating particles of the medium do not travel these distances. In view of this, what can we say about the nature of light: does light travel as a stream of particles away from its source, or does light travel in the form of waves that spread outward from the source?

900

Historically, this question has turned out to be a difficult one. For one thing, light does not reveal itself in any obvious way as being made up of tiny particles; nor do we see tiny light waves passing by as we do water waves. The evidence seemed to favor first one side and then the other until about 1830, when most physicists had accepted the wave theory. By the end of the nineteenth century, light was considered to be an electromagnetic wave (Chapter 31). In the early twentieth century, light was shown to have a particle nature as well, as we shall discuss in Chapter 37. We now speak of the wave-particle duality of light. The wave theory of light remains valid and has proved very successful. We now investigate the evidence for the wave theory and how it has been used to explain a wide range of phenomena.

34-1 Waves versus Particles; Huygens' Principle and Diffraction

The Dutch scientist Christian Huygens (1629-1695), a contemporary of Newton, proposed a wave theory of light that had much merit. S till useful today is a technique Huygens developed for predicting the future position of a wave front when an earlier position is known. By a wave front, we mean all the points along a two- or three-dimensional wave that form a wave crest— what we simply call a “wave” as seen on the ocean. Wave fronts are perpendicular to rays as we already discussed in Chapter 15 (Fig. 15-20). Huygens’ principle can be stated as follows: Every poin t on a w ave front can be considered as a source o f tiny wavelets that spread out in the forw ard direction at the speed o f the w ave itself The new w ave front is the envelope o f all the wavelets— that is, the tangent to all o f them.

As a simple example of the use of Huygens’ principle, consider the wave front AB in Fig. 34-1, which is traveling away from a source S. We assume the medium is isotropic— that is, the speed v of the waves is the same in all directions. To find the wave front a short time t after it is at AB, tiny circles are drawn with radius r = vt. The centers of these tiny circles are blue dots on the original wave front AB, and the circles represent Huygens’ (imaginary) wavelets. The tangent to all these wavelets, the curved line CD, is the new position of the wave front.

Huygens’ principle is particularly useful for analyzing what happens when waves impinge on an obstacle and the wave fronts are partially interrupted. Huygens’ principle predicts that waves bend in behind an obstacle, as shown in Fig. 34-2. This is just what water waves do, as we saw in Chapter 15 (Figs. 15-31 and 15-32). The bending of waves behind obstacles into the “ shadow region” is known as diffraction. Since diffraction occurs for waves, but not for particles, it can serve as one means for distinguishing the nature of light.

Note, as shown in Fig. 34-2, that diffraction is most prominent when the size of the opening is on the order of the wavelength of the wave. If the opening is much larger than the wavelength, diffraction goes unnoticed.

Does light exhibit diffraction? In the mid-seventeenth century, the Jesuit priest Francesco Grimaldi (1618-1663) had observed that when sunlight entered a darkened room through a tiny hole in a screen, the spot on the opposite wall was larger than would be expected from geometric rays. He also observed that the border of the image was not clear but was surrounded by colored fringes. Grimaldi attributed this to the diffraction of light.

The wave model of light nicely accounts for diffraction, and we discuss diffraction in detail in the next Chapter. But the ray model (Chapter 32) cannot account for diffraction, and it is important to be aware of such limitations to the ray model. Geometric optics using rays is successful in a wide range of situations only because normal openings and obstacles are much larger than the wavelength of the light, and so relatively little diffraction or bending occurs.

Source

S

FIGURE 34-1 Huygens’ principle, used to determine wave front CD when wave front AB is given.

FIGURE 34-2 Huygens’ principle is consistent with diffraction(a) around the edge of an obstacle,(b) through a large hole, (c) through a small hole whose size is on the order of the wavelength of the wave.

(c)

SECTION 34-1 Waves versus Particles; Huygens' Principle and Diffraction 901

34—2 Huygens' Principle and the Law of Refraction

FIGURE 34-3 Refraction explained, using Huygens’ principle. Wave fronts are perpendicular to the rays.

902 CHAPTER 34

The laws of reflection and refraction were well known in Newton’s time. The law of reflection could not distinguish between the two theories we just discussed: waves versus particles. For when waves reflect from an obstacle, the angle of incidence equals the angle of reflection (Fig. 15-21). The same is true of particles— think of a tennis ball without spin striking a flat surface.

The law of refraction is another matter. Consider a ray of light entering a medium where it is bent toward the normal, as when traveling from air into water. As shown in Fig. 34-3, this bending can be constructed using Huygens’ principle if we assume the speed of light is less in the second medium (v2 < vJ. In time t, point B on wave front AB (perpendicular to the incoming ray) travels a distance vx t to reach point D. Point A on the wave front, traveling in the second medium, goes a distance v2t to reach point C, and v2t < vxt. Huygens’ principle is applied to points A and B to obtain the curved wavelets shown at C and D. The wave front is tangent to these two wavelets, so the new wave front is the line CD. Hence the rays, which are perpendicular to the wave fronts, bend toward the normal if v2 < vx, as drawn.

Newton favored a particle theory of light which predicted the opposite result, that the speed of light would be greater in the second medium (v2 > vJ. Thus the wave theory predicts that the speed of light in water, for example, is less than in air; and Newton’s particle theory predicts the reverse. An experiment to actually measure the speed of light in water was performed in 1850 by the French physicist Jean Foucault, and it confirmed the wave-theory prediction. By then, however, the wave theory was already fully accepted, as we shall see in the next Section.

Snell’s law of refraction follows directly from Huygens’ principle, given that the speed of light v in any medium is related to the speed in a vacuum, c, and the index of refraction, n, by Eq. 32-4: that is, v = c/n. From the Huygens’ construction of Fig. 34-3, angle ADC is equal to d2 and angle BAD is equal to 61. Then for the two triangles that have the common side AD, we have

vxt v2t= A D ’ 2 = AD

We divide these two equations and obtain sin 61 vx sin 02 v2

Then, by Eq. 32-4 vx = c/nx and v2 = c/n2, so we have

nx sin 6X = n2 sin 02,which is Snell’s law of refraction, Eq. 32-5. (The law of reflection can be derived from Huygens’ principle in a similar way: see Problem 1 at the end of this Chapter.)

When a light wave travels from one medium to another, its frequency does not change, but its wavelength does. This can be seen from Fig. 34-3, where each of the blue lines representing a wave front corresponds to a crest (peak) of the wave. Then

A2 v2 t v2 nx Ax vx t vx n2

where, in the last step, we used Eq. 32-4, v = c/n. I f medium 1 is a vacuum (or air), so nx = 1, vx = c, and we call \ x simply A, then the wavelength in another medium of index of refraction n (= n2) w ill be

K = ~ (34-1)This result is consistent with the frequency / being unchanged no matter what medium the wave is traveling in, since c = /A .

EXERCISE A A light beam in air with wavelength = 500 nm, frequency = 6.0 X 1014 Hz, and speed = 3.0 X 108m/s goes into glass which has an index of refraction = 1.5. What are the wavelength, frequency, and speed of the light in the glass?

(a)

Wave fronts can be used to explain how mirages are produced by refraction of light. For example, on a hot day motorists sometimes see a mirage of water on the highway ahead of them, with distant vehicles seemingly reflected in it (Fig. 34-4a). On a hot day, there can be a layer of very hot air next to the roadway (made hot by the Sun beating on the road). Hot air is less dense than cooler air, so the index of refrac­tion is slightly lower in the hot air. In Fig. 34-4b, we see a diagram of light coming from one point on a distant car (on the right) heading left toward the observer. Wave fronts and two rays (perpendicular to the wave fronts) are shown. Ray A heads directly at the observer and follows a straight-line path, and represents the normal view of the distant car. Ray B is a ray initially directed slightly downward, but it bends slightly as it moves through layers of air of different index of refraction. The wave fronts, shown in blue in Fig. 34-4b, move slightly faster in the layers of air nearer the ground. Thus ray B is bent as shown, and seems to the observer to be coming from below (dashed line) as if reflected off the road. Hence the mirage.

34—3 Interference—Young's Double-Slit Experiment

In 1801, the Englishman Thomas Young (1773-1829) obtained convincing evidence for the wave nature of light and was even able to measure wavelengths for visible light. Figure 34-5a shows a schematic diagram of Young’s famous double-slit experiment.

FIGURE 34-5 (a) Young’s double-slit experiment, (b) If light consists of particles, we would expect to see two bright lines on the screen behind the slits, (c) In fact, many lines are observed. The slits and their separation need to be very thin.

Sun’srays

^1

---S2 B S(a) Viewing screen (b) Viewing screen

(particle theory prediction)

(c) Viewing screen (actual)

To have light from a single source, Young used the Sun passing through a very narrow slit in a window covering. This beam of parallel rays falls on a screen containing two closely spaced slits, Si and S2. (The slits and their separation are very narrow, not much larger than the wavelength of the light.) I f light consists of tiny particles, we might expect to see two bright lines on a screen placed behind the slits as in (b). But instead a series of bright lines are seen, as in (c). Young was able to explain this result as a wave-interference phenomenon.

To understand why, we consider the simple situation of plane waves of light of a single wavelength— called monochromatic, meaning “ one color” — falling on the two slits as shown in Fig. 34-6. Because of diffraction, the waves leaving the two small slits spread out as shown. This is equivalent to the interference pattern produced when two rocks are thrown into a lake (Fig. 15-23), or when sound from two loudspeakers interferes (Fig. 16-15). Recall Section 15-8 on wave interference.

FIGURE 34-4 (a) A highway mirage, (b) Drawing (greatly exaggerated) showing wave fronts and rays to explain highway mirages. Note how sections of the wave fronts near the ground move faster and so are farther apart.

0 P H Y S I C S A P P L I E DHighway mirages

FIGURE 34-6 If light is a wave, light passing through one of two slits should interfere with light passing through the other slit.

Rays

Wave fronts

Direct rayRay directed

slightly downwardObserver

SECTION 34-3 Interference-Young's Double-Slit Experiment 903

Bright (constructive interference)

- l - Screen(a) 0 = 0° (d)

FIGURE 34-7 How the wave theory explains the pattern of lines seen in the double-slit experiment.(a) At the center of the screen the waves from each slit travel the same distance and are in phase.(b) At this angle 0, the lower wave travels an extra distance of one whole wavelength, and the waves are in phase; note from the shaded triangle that the path difference equals d sin 0. (c) For this angle 0, the lower wave travels an extra distance equal to one-half wavelength, so the two waves arrive at the screen fully out of phase, (d) A more detailed diagram showing the geometry for parts (b) and (c).

FIGURE 34-8 Two traveling waves are shown undergoing(a) constructive interference,(b) destructive interference. (See also Section 15-8.)

Wa A -fm

w\AA;

(b)

To see how an interference pattern is produced on the screen, we make use of Fig. 34-7. Waves of wavelength A are shown entering the slits Sx and S2, which are a distance d apart. The waves spread out in all directions after passing through the slits (Fig. 34-6), but they are shown only for three different angles 0. In Fig. 34-7a, the waves reaching the center of the screen are shown (0 = 0°). The waves from the two slits travel the same distance, so they are in phase: a crest of one wave arrives at the same time as a crest of the other wave. Hence the amplitudes of the two waves add to form a larger amplitude as shown in Fig. 34-8a. This is constructive interference, and there is a bright area at the center of the screen. Constructive interference also occurs when the paths of the two rays differ by one wavelength (or any whole number of wavelengths), as shown in Fig. 34-7b; also here there w ill be brightness on the screen. But if one ray travels an extra distance of one-half wavelength (or §A, f A, and so on), the two waves are exactly out of phase when they reach the screen: the crests of one wave arrive at the same time as the troughs of the other wave, and so they add to produce zero amplitude (Fig. 34-8b). This is destructive interference, and the screen is dark, Fig. 34-7c. Thus, there w ill be a series of bright and dark lines (or fringes) on the viewing screen.

To determine exactly where the bright lines fall, first note that Fig. 34-7 is some­what exaggerated; in real situations, the distance d between the slits is very small compared to the distance £ to the screen. The rays from each slit for each case w ill therefore be essentially parallel, and 0 is the angle they make with the horizontal as shown in Fig. 34-7d. From the shaded right triangles shown in Figs. 34-7b and c, we can see that the extra distance traveled by the lower ray is d sin 6 (seen more clearly in Fig. 34-7d). Constructive interference w ill occur, and a bright fringe w ill appear on the screen, when the path difference, d sin 0 , equals a whole number of wavelengths:

d sin 9 = raA, ra = 0,1,2,constructiveinterference

(bright)(34-2a)

The value of ra is called the order of the interference fringe. The first order (ra = 1 ), for example, is the first fringe on each side of the central fringe (which is at 6 = 0, ra = 0 ). Destructive interference occurs when the path difference d sin 0 is |A ,|A , and so on:

d sin 6 = (ra + |)A, ra = 0,1,2,destructive

interference(dark)

(34-2b)

The bright fringes are peaks or maxima of light intensity, the dark fringes are minima.

sl ~s2 = d sin0

904 CHAPTER 34 The Wave Nature of Light; Interference

The intensity of the bright fringes is greatest for the central fringe (ra = 0) and decreases for higher orders, as shown in Fig. 34-9. How much the intensity decreases with increasing order depends on the width of the two slits.

CONCEPTUAL EXAMPLE 54-1 I Interference pattern lines, (a) W ill there bean infinite number of points on the viewing screen where constructive and destruc­tive interference occur, or only a finite number of points? (b) Are neighboring points of constructive interference uniformly spaced, or is the spacing between neighboring points of constructive interference not uniform?

RESPONSE (a) When you look at Eqs. 34-2a and b you might be tempted to say, given the statement ra = 0 , 1, 2 , beside the equations, that there are an infinite number of points of constructive and destructive interference. However, recall that sin 9 cannot exceed 1. Thus, there is an upper lim it to the values of ra that can be used in these equations. For Eq. 34-2a, the maximum value of ra is the integer closest in value but smaller than d/X. So there are & finite number of points of constructive and destructive interference no matter how large the screen, (b) The spacing between neighboring points of constructive or destructive interference is not uniform: The spacing gets larger as 6 gets larger, and you can verify this statement mathematically. For small values of 6 the spacing is nearly uniform as you w ill see in Example 34-2.

Line spacing for double-slit interference. A screen containing two slits 0.100 mm apart is 1.20 m from the viewing screen. Light of wavelength A = 500 nm falls on the slits from a distant source. Approximately how far apart w ill adjacent bright interference fringes be on the screen?

APPROACH The angular position of bright (constructive interference) fringes is found using Eq. 34-2a. The distance between the first two fringes (say) can be found using right triangles as shown in Fig. 34-10.SOLUTION Given d = 0.100 mm = 1.00 X HT4m, A = 500 X 10“9m, and £ = 1.20 m, the first-order fringe (ra = 1 ) occurs at an angle 6 given by

m \ (1)(500 X l(T 9m)sin#! = — — = — — — ---- —— ------ = 5.00 X 10 .

d 1.00 X 10“4 m

This is a very small angle, so we can take sin 0 « 0, with 0 in radians. The first-order fringe w ill occur a distance xx above the center of the screen (see Fig. 34-10), given by x j l = tan#! » 01? so

~ £0X = (1.20m)(5.00 X IO-3) = 6.00 mm.

The second-order fringe (ra = 2) w ill occur at

2ax2 ~ 102 = £ —~ = 12.0 mm

dS

above the center, and so on. Thus the lower order fringes are 6.00 mm apart. s NOTE The spacing between fringes is essentially uniform until the approximation sin 0 ^ 0 is no longer valid.

CONCEPTUAL EXAMPLE 34-3 I Changing the wavelength, (a) What happensto the interference pattern shown in Fig. 34-10, Example 34-2, if the incident light (500 nm) is replaced by light of wavelength 700 nm? (b) What happens instead if the wavelength stays at 500 nm but the slits are moved farther apart?

RESPONSE (a) When A increases in Eq. 34-2a but d stays the same, then the angle 0 for bright fringes increases and the interference pattern spreads out.(b) Increasing the slit spacing d reduces 0 for each order, so the lines are closer together.

Constructive interference(a)

m =3 2 1 0 1 2 3

m= 2 1 0 0 1 2 3Destructive interference

(b)

FIGURE 34-9 (a) Interference fringes produced by a double-slit experiment and detected by photographic film placed on the viewing screen. The arrow marks the central fringe, (b) Graph of the intensity of light in the interference pattern. Also shown are values of m for Eq. 34-2a (constructive interference) and Eq. 34-2b (destructive interference).

FIGURE 34-10 Examples 34-2 and 34-3. For small angles 6 (give 6 in radians), the interference fringes occur at distance x = $£ above the center fringe (m = 0); and xi are for the first-order fringe (m = 1),02 and x2 are for m = 2.

SECTION 34-3 Interference-Young's Double-Slit Experiment 905

White

II I Ih-2 .0 mm-H

H------ 3.5 mm

FIGURE 34-11 First-order fringes are a full spectrum, like a rainbow. Also Example 34-4.

From Eqs. 34-2 we can see that, except for the zeroth-order fringe at the center, the position of the fringes depends on wavelength. Consequently, when white light falls on the two slits, as Young found in his experiments, the central fringe is white, but the first- (and higher-) order fringes contain a spectrum of colors like a rainbow; 0 was found to be smallest for violet light and largest for red (Fig. 34-11). By measuring the position of these fringes, Young was the first to determine the wavelengths of visible light (using Eqs. 34-2). In doing so, he showed that what distinguishes different colors physically is their wavelength (or frequency), an idea put forward earlier by Grimaldi in 1665.

FIGURE 34-10 (Repeated.)For small angles 0 (give 6 in radians), the interference fringes occur at distance x = di above the center fringe (ra = 0); and x\ are for the first-order fringe (ra = 1),02 and x2 are for ra = 2.

EXAMPLE 34-4 Wavelengths from double-slit interference. White light passes through two slits 0.50 mm apart, and an interference pattern is observed on a screen 2.5 m away. The first-order fringe resembles a rainbow with violet and red light at opposite ends. The violet light is about 2.0 mm and the red 3.5 mm from the center of the central white fringe (Fig. 34-11). Estimate the wavelengths for the violet and red light.

APPROACH We find the angles for violet and red light from the distances given and the diagram of Fig. 34-10. Then we use Eq. 34-2a to obtain the wavelengths. Because 3.5 mm is much less than 2.5 m, we can use the small-angle approximation. SOLUTION We use Eq. 34-2a with ra = 1 and sin 0 ~ tan 6 ~ 6. Then for violet light, x = 2.0 mm, so (see also Fig. 34-10)

A =d sin0

ra md x ra i

5.0 X 10-4 m \ / 2.0 X 10“31

m2.5 m

= 4.0 X 10_7m

or 400 nm. For red light, x = 3.5 mm, so

_ d x m l ~

5.0 X 10_4m \ / 3.5 X 10“31

m2.5 m

= 7.0 X 10-7 m = 700 nm.

CoherenceThe two slits in Fig. 34-7 act as if they were two sources of radiation. They are called coherent sources because the waves leaving them have the same wavelength and frequency, and bear the same phase relationship to each other at all times. This happens because the waves come from a single source to the left of the two slits in Fig. 34-7, splitting the original beam into two. An interference pattern is observed only when the sources are coherent. I f two tiny lightbulbs replaced the two slits, an interference pattern would not be seen. The light emitted by one lightbulb would have a random phase with respect to the second bulb, and the screen would be more or less uniformly illuminated. Two such sources, whose output waves have phases that bear no fixed relationship to each other over time, are called incoherent sources.

*34—4 Intensity in the Double-Slit Interference Pattern

We saw in Section 34-3 that the interference pattern produced by the coherent light from two slits, Sx and S2 (Figs. 34-7 and 34-9), produces a series of bright and dark fringes. I f the two monochromatic waves of wavelength A are in phase at the slits, the maxima (brightest points) occur at angles 0 given by (Eqs. 34-2)

d sin 0 = raA,

and the minima (darkest points) when

dsinO = (ra + !)A ,

where ra is an integer (m = 0 , 1, 2 , • • •).

906 CHAPTER 34 The Wave Nature of Light; Interference

We now determine the intensity of the light at all points in the pattern, assuming that if either slit were covered, the light passing through the other would diffract sufficiently to illuminate a large portion of the screen uniformly. The intensity I of the light at any point is proportional to the square of its wave amplitude (Section 15-3). Treating light as an electromagnetic wave, I is propor­tional to the square of the electric field E (or to the magnetic field B, Section 31-8):I oc E 2. The electric field E at any point P (see Fig. 34-12) w ill be the sum of the electric field vectors of the waves coming from each of the two slits, E1 and E2. Since E1 and E2 are essentially parallel (on a screen far away compared to the slit separation), the magnitude of the electric field at angle 6 (that is, at point P) w ill be

Ee = E1 + E2.

Both E1 and E2 vary sinusoidally with frequency / = c/A, but they differ in phase, depending on their different travel distances from the slits. The electric field at P can then be written for the light from each of the two slits, using cd = 2irf, as

Ei = Ew sm<otE2 = E20sm((i)t + 8)

where Eio and E2o are their respective amplitudes and 8 is the phase difference. The value of 8 depends on the angle 0 , so let us now determine 8 as a function of 0.

A t the center of the screen (point 0), 5 = 0. If the difference in path length from P to S1 and S2 is d sin 0 = A/2, the two waves are exactly out of phase so 8 = ir (or 180°). I f d sin 0 = A, the two waves differ in phase by 8 = 2tt. In general, then, 8 is related to 0 by

8 _ d sin 02tt A

8 = —̂ ~d sin0. (34-4)A

To determine Ed = E1 + E2, we add the two scalars E1 and E2 which are sine functions differing by the phase 8. One way to determine the sum of E1 and E2 is to use a phasor diagram. (We used this technique before, in Chapter 30.) As shown in Fig. 34-13, we draw an arrow of length Eio to represent the amplitude of Ex (Eq. 34-3); and the arrow of length E2q > which we draw to make a fixed angle 8 with -̂ 10 > represents the amplitude of E2. When the diagram rotates at angular frequency to about the origin, the projections of £io and 2̂0 on the vertical axis represent E1 and E2 as a function of time (see Eq. 34-3). We let Ego be the “vector” sumf of Ew and E2o; Eeo is the amplitude of the sum Ee = El + E2, and the projection of Eeo on the vertical axis is just Ee . I f the two slits provide equal illumination, so that E10 = E20 = E0, then from symmetry in Fig. 34-13, the angle (j> = 8/ 2, and we can write

+ (34-5a)

From Fig. 34-13 we can also see that

Eeo = 2E0 co s <f) = 2E0 co s —. (34-5b)

Combining Eqs. 34-5a and b, we obtain

Ee = 2E0 co s ^ sin + ^ ) , (34-5c)

where 8 is given by Eq. 34-4.

fWe are not adding the actual electric field vectors; instead we are using the “phasor” technique to add the amplitudes, taking into account the phase difference of the two waves.

S2

Screen

FIGURE 34-12 Determining the intensity in a double-slit interference pattern. Not to scale: in fact £ » d, and the two rays become essentially parallel.

FIGURE 34-13 Phasor diagram for double-slit interference pattern.

*SECTION 34-4 Intensity in the Double-Slit Interference Pattern 907

We are not really interested in Ee as a function of time, since for visible light the frequency (1014 to 1015 Hz) is much too high to be noticeable. We are interested in the average intensity, which is proportional to the amplitude squared, Eg0 • We

point P at an angle 0 to the horizontal. We let 70 be the intensity at point O, the center of the screen, where 6 = 8 = 0, so 70 oc (E10 + E20)2 = (2E0)2. Then

these two points, so

where we used Eq. 34-5b. Thus the intensity Ie at any point is related to the maximum intensity at the center of the screen by

where 8 was given by Eq. 34-4. This is the relation we sought.From Eq. 34-6 we see that maxima occur where cos 8 /2 = ± 1, which

corresponds to 5 = 0, 2tt, 4tt, • • •. From Eq. 34-4, 8 has these values when

These are the same results we obtained in Section 34-3. But now we know not only the position of maxima and minima, but from Eq. 34-6 we can determine the intensity at all points.

In the usual situation where the distance £ to the screen from the slits is large compared to the slit separation d (£ » d), if we consider only points P whose distance y from the center (point O) is small compared to £ (y « £)— see Fig. 34-12— then

EXERCISE B What are the values for the intensity Ie when (a) y = 0, (b) y = \£/4d, and(c) y = X£/2dl

The intensity Ie as a function of the phase difference 8 is plotted in Fig. 34-14. In the approximation of Eq. 34-7, the horizontal axis could as well be y, the position on the screen.

now drop the word “ average,” and we let Ie (Ie oc Eg) be the intensity at any

the ratio Ie/ I0 is equal to the ratio of the squares of the electric-field amplitudes at

(34-6)

d sin0 = raA, ra = 0,1,2, •••.

Minima occur where 8 = tt, 3tt, 5-77, • • •, which corresponds to

dsin0 = (ra + !)A, ra = 0,1,2, •••.

From this it follows (see Eq. 34-4) that

Equation 34-6 then becomes

[y « £, d « £] (34-7)

908 CHAPTER 34 The Wave Nature of Light; Interference

h2 coherent sources

2 incoherent sources

-577 -473- -377 -277 -77 0

H-------1-------1-------1-------1-------h-5AI -2A l -3A£ -U -XJ_ 0 2d d 2d d 2d

77 277 377 477 577

—\------ 1------ 1------ 1------ \+y (if y « i)Al Al 3AI 2Al 5Al 2d d 2d d Id

FIGURE 34-14 Intensity / as a function of phase difference 8 and position on screen y (assumingy « *)•

The intensity pattern expressed in Eqs. 34-6 and 34-7, and plotted in Fig. 34-14, shows a series of maxima of equal height, and is based on the assumption that each slit (alone) would illuminate the screen uniformly. This is never quite true, as we shall see when we discuss diffraction in the next Chapter. We w ill see that the center maximum is strongest and each succeeding maximum to each side is a little less strong.

Antenna intensity. Two radio antennas are located close to each other as shown in Fig. 34-15, separated by a distance d. The antennas radiate in phase with each other, emitting waves of intensity I0 at wavelength A.(a) Calculate the net intensity as a function of 6 for points very far from the antennas. (b) For d = A, determine I and find in which directions I is a maximum and a minimum, (c) Repeat part (b ) when d = A/2.

APPROACH This setup is similar to Young’s double-slit experiment.SOLUTION (a) Points of constructive and destructive interference are still given by Eqs. 34-2a and b, and the net intensity as a function of 6 is given by Eq. 34-6.(b) We let d = A in Eq. 34-6, and find for the intensity,

I = /ocos2(7rs in 0 ).

I is a maximum, equal to 70, when sin 0 = 0, 1, or -1 , meaning 6 = 0, 90°, 180°, and 270°. I is zero when sin 6 = \ and for which 6 = 30°, 150°, 210°, and 330°.(c) For d = A/2, I is maximized for 6 = 0 and 180°, and minimized for 90° and 270°.

34—5 Interference in Thin FilmsInterference of light gives rise to many everyday phenomena such as the bright colors reflected from soap bubbles and from thin oil or gasoline films on water,Fig. 34-16. In these and other cases, the colors are a result of constructive interfer­ence between light reflected from the two surfaces of the thin film.

FIGURE 34-16 Thin film interference patterns seen in (a) a soap bubble, (b) a thin film of soapy water, and (c) a thin layer of oil on wet pavement.

FIGURE 34-15 Example 34-5. The two dots represent the antennas.

(a)

SECTION 34-5 Interference in Thin Films 909

AirOil

Water

FIGURE 34-17 Light reflected from the upper and lower surfaces of a thin film of oil lying on water. This analysis assumes the light strikes the surface nearly perpendicularly, but is shown here at an angle so we can display each ray.

To see how this thin-film interference happens, consider a smooth surface of water on top of which is a thin uniform layer of another substance, say anoil whose index of refraction is less than that of water (we’ll see why we assume this in a moment); see Fig. 34-17. Assume for the moment that the incident light is of a single wavelength. Part of the incident light is reflected at A on the top surface, and part of the light transmitted is reflected at B on the lower surface. The part reflected at the lower surface must travel the extra distance ABC. I f this path difference ABC equals one or a whole number of wavelengths in the film (A„), the two waves w ill reach the eye in phase and interfere constructively. Hence the region AC on the surface film w ill appear bright. But if ABC equals ^An,§An, and so on, the two waves w ill be exactly out of phase and destructive interference occurs: the area AC on the film w ill show no reflection— it w ill be dark (or better, transparent to the dark material below). The wavelength Xn is the wavelength in the film: Xn = X/n, where n is the index of refraction in the film and A is the wavelength in vacuum. See Eq. 34-1.

When white light falls on such a film , the path difference ABC w ill equal Xn (or mXn, with m = an integer) for only one wavelength at a given viewing angle. The color corresponding to A (A in air) w ill be seen as very bright. For light viewed at a slightly different angle, the path difference ABC w ill be longer or shorter and a different color w ill undergo constructive interference. Thus, for an extended (nonpoint) source emitting white light, a series of bright colors w ill be seen next to one another. Variations in thickness of the film w ill also alter the path difference ABC and therefore affect the color of light that is most strongly reflected.

EXERCISE C Return to the Chapter-Opening Question, page 900, and answer it again now. Try to explain why you may have answered differently the first time.

When a curved glass surface is placed in contact with a flat glass surface, Fig. 34-18, a series of concentric rings is seen when illuminated from above by either white light (as shown) or by monochromatic light. These are called Newton’s ringst and they are due to interference between waves reflected by the top and bottom surfaces of the very thin air gap between the two pieces of glass. Because this gap (which is equivalent to a thin film ) increases in width from the central contact point out to the edges, the extra path length for the lower ray (equal to BCD) varies; where it equals 0, \X , A, §A, 2A, and so on, it corresponds to constructive and destructive interference; and this gives rise to the series of bright colored circles seen in Fig. 34-18b. The color you see at a given radius corresponds to constructive interference; at that radius, other colors partially or fully destructively interfere. ( If monochromatic light is used, the rings are alternately bright and dark.)

f Although Newton gave an elaborate description of them, they had been first observed and described by his contemporary, Robert Hooke.

FIGURE 34-18 Newton’s rings.(a) Light rays reflected from upper and lower surfaces of the thin air gap can interfere.(b) Photograph of interference patterns using white light.

910 CHAPTER 34 The Wave Nature of Light; Interference

The point of contact of the two glass surfaces (A in Fig. 34-18a) is dark in Fig. 34-18b. Since the path difference is zero here, our previous analysis would suggest that the waves reflected from each surface are in phase and so this central area ought to be bright. But it is dark, which tells us something else is happening here: the two waves must be completely out of phase. This can happen if one of the waves, upon reflection, flips over— a crest becomes a trough— see Fig. 34-19. We say that the reflected wave has undergone a phase change of 180°, or of half a wave cycle. Indeed, this and other experiments reveal that, at normal incidence,

a beam of light reflected by a material with index of refraction greater than that of the material in which it is traveling, changes phase by 180° or \ cycle;

see Fig. 34-19. This phase change acts just like a path difference of \ A. I f the index of refraction of the reflecting material is less than that of the material in which the light is traveling, no phase change occurs.f

Thus the wave reflected at the curved surface above the air gap in Fig. 34-18a undergoes no change in phase. But the wave reflected at the lower surface, where the beam in air strikes the glass, undergoes a |-cycle phase change, equivalent to a \ A path difference. Thus the two waves reflected near the point of contact A of the two glass surfaces (where the air gap approaches zero thickness) w ill be a half cycle (or 180°) out of phase, and a dark spot occurs. Bright colored rings w ill occur when the path difference is \X, § A, and so on, because the phase change at one surface effectively adds a path difference of \X (=§ cycle). (If monochromatic light is used, the bright Newton’s rings w ill be separated by dark bands which occur when the path difference BCD in Fig. 34-18a is equal to an integral number of wavelengths.)

Returning for a moment to Fig. 34-17, the light reflecting at both interfaces, a ir-o il and oil-water, underwent a phase change of 180° equivalent to a path difference of \X, since we assumed nwater > noil > nair; since the phase changes were equal, they didn’t affect our analysis.

■ Thin film of air, wedge-shaped. A very fine wire7.35 X IO-3 mm in diameter is placed between two flat glass plates as in Fig. 34-20a. Light whose wavelength in air is 600 nm falls (and is viewed) perpendicular to the plates and a series of bright and dark bands is seen, Fig. 34-20b. How many light and dark bands w ill there be in this case? W ill the area next to the wire be bright or dark?

APPROACH We need to consider two effects: (1) path differences for rays reflecting from the two close surfaces (thin wedge of air between the two glass plates), and (2) the ^-cycle phase change at the lower surface (point E in Fig. 34-20a), where rays in air can enter glass. Because of the phase change at the lower surface, there w ill be a dark band (no reflection) when the path difference is 0, A, 2A, 3A, and so on. Since the light rays are perpendicular to the plates, the extra path length equals 21, where t is the thickness of the air gap at any point. SOLUTION Dark bands w ill occur where

21 = mX, m = 0,1,2, •••.

Bright bands occur when 21 = (m + )̂A, where m is an integer. A t the position of the wire, t = 7.35 X 10_6m. A t this point there w ill be 2t/X = (2)(7.35 X 10-6 m)/(6.00 X 10“7 m) = 24.5 wavelengths. This is a “half integer,” so the area next to the wire w ill be bright. There w ill be a total of 25 dark lines along the plates, corresponding to path lengths of 0A, IA, 2A, 3A,..., 24A, including the one at the point of contact A (m = 0). Between them, there w ill be 24 bright lines plus the one at the end, or 25.NOTE The bright and dark bands w ill be straight only if the glass plates are extremely flat. I f they are not, the pattern is uneven, as in Fig. 34-20c. Thus we see a very precise way of testing a glass surface for flatness. Spherical lens surfaces can be tested for precision by placing the lens on a flat glass surface and observing Newton’s rings (Fig. 34-18b) for perfect circularity.

n\

(a)

\f\fw

n\

\7V v/V

n2>nl

(b)

FIGURE 34-19 (a) Reflected ray changes phase by 180° or \ cycle if n2 > n i , but (b) does not if n2 < n\.FIGURE 34-20 (a) Light rays reflected from the upper and lower surfaces of a thin wedge of air interfere to produce bright and dark bands.(b) Pattern observed when glass plates are optically flat; (c) pattern when plates are not so flat. See Example 34-6.&

*

(°)g ) P H Y S I C S A P P L I E DTesting glass for flatness

trThis result can be derived from Maxwell’s equations. It corresponds to the reflection of a wave trav­eling along a cord when it reaches the end; as we saw in Fig. 15-18, if the end is tied down, the wave changes phase and the pulse flips over, but if the end is free, no phase change occurs. SECTION 34-5 911

n = 1.35

Incident ray. Reflected rays’

Outside air n = 1.00

Bubble interior n = 1.00

-H t h-FIGURE 34-21 Example 34-7.The incident and reflected rays are assumed to be perpendicular to the bubble’s surface. They are shown at a slight angle so we can distinguish them.

/?\ CAUTION____________A formula is not enough: you must also

check for phase changes at surfaces

FIGURE 34-16b (Repeated.)

If the wedge between the two glass plates of Example 34-6 is filled with some transparent substance other than air— say, water— the pattern shifts because the wavelength of the light changes. In a material where the index of refraction is n, the wavelength is An = A/n , where A is the wavelength in vacuum (Eq. 34-1). For instance, if the thin wedge of Example 34-6 were filled with water, then kn = 600nm/1.33 = 450 nm; instead of 25 dark lines, there would be 33.

When white light (rather than monochromatic light) is incident on the thin wedge of air in Figs. 34-18a or 34-20a, a colorful series of fringes is seen because constructive interference occurs for different wavelengths in the reflected light at different thicknesses along the wedge.

A soap bubble (Fig. 34-16a and Chapter-Opening Photo) is a thin spherical shell (or film ) with air inside. The variations in thickness of a soap bubble film gives rise to bright colors reflected from the soap bubble. (There is air on both sides of the bubble film .) Similar variations in film thickness produce the bright colors seen reflecting from a thin layer of oil or gasoline on a puddle or lake (Fig. 34-16c). Which wavelengths appear brightest also depends on the viewing angle.

■ Thickness of soap bubble skin. A soap bubble appearsgreen (A = 540 nm) at the point on its front surface nearest the viewer. What is the smallest thickness the soap bubble film could have? Assume n = 1.35. APPROACH Assume the light is reflected perpendicularly from the point on a spherical surface nearest the viewer, Fig. 34-21. The light rays also reflect from the inner surface of the soap bubble film as shown. The path difference of these two reflected rays is 21, where t is the thickness of the soap film . Light reflected from the first (outer) surface undergoes a 180° phase change (index of refraction of soap is greater than that of air), whereas reflection at the second (inner) surface does not. To determine the thickness t for an interference maximum, we must use the wavelength of light in the soap (n = 1.35).SOLUTION The 180° phase change at only one surface is equivalent to a §A path difference. Therefore, green light is bright when the minimum path difference equals ^An. Thus, 21 = A/2n, so

A (540 nm)= 100 nm.

912 CHAPTER 34

(4)(1.35)This is the smallest thickness; but the green color is more likely to be seen at the next thickness that gives constructive interference, 21 = 3 \/2 n , because other colors would be more fully cancelled by destructive interference. The more likely thickness is 3A/4n = 300 nm, or even 5A/4n = 500 nm. Note that green is seen in air, so A = 540 nm (not A/«).

* Colors in a Thin Soap FilmThe thin film of soapy water shown in Fig. 34-16b (repeated here) has stood vertically for a long time so that gravity has pulled much of the soapy water toward the bottom. The top section is so thin (perhaps 30 nm thick) that light reflected from the front and back surfaces have almost no path difference. Thus the 180° phase change at the front surface assures that the two reflected waves are 180° out of phase for all wavelengths of visible light. The white light incident on this thin film does not reflect at the top part of the film. Thus the top is transparent, and we see the background which is black.

Below the black area at the top, there is a thin blue line, and then a white band. The film thickness is perhaps 75 to 100 nm, so the shortest wavelength (blue) light begins to partially interfere constructively; but just below, where the thickness is slightly greater (100 nm), the path length is reasonably close to A/2 for much of the spectrum and we see white or silver. (Why? Recall that red starts at 600 nm in air; so most colors in the spectrum lie between 450 nm and 600 nm in air; but in water the wavelengths are n = 1.33 times smaller, 340 nm to 450 nm, so a 100 nm thickness is a 200 nm path length, not far from A/2 for most colors.) Immediately below the white band we see a brown band (around 200 nm in thickness) where selected wavelengths (not all) are close to exactly A and those colors destructively interfere, leaving only a few colors to partially interfere constructively, giving us murky brown.

9.0

B£

Farther down, with increasing thickness t, a path length 21 = 510 nm corresponds nicely to §A for blue, but not for other colors, so we see blue (§A path difference plus \ A phase change = constructive interference). Other colors experience constructive interference (at §A and then at f A) at still greater thicknesses, so we see a series of separated colors something like a rainbow.

In the soap bubble of our Chapter-Opening Photo (p. 900), similar things happen: at the top (where the film is thinnest) we see black and then silver-white, just as within the loop shown in Fig. 34-16b. And examine the oil film on wet pavement shown in Fig. 34-16c (repeated here); the oil film is thickest at the center and thins out toward the edges. Notice the whitish outer ring where most colors constructively interfere, which would suggest a thickness on the order of 100 nm as discussed above for the white band in the soap film. Beyond the outer white band of the oil film , Fig. 34-16c, there is still some oil, but the film is so thin that reflected light from upper and lower surfaces destructively interfere and you can see right through this very thin oil film.

FIGURE 34-16C (Repeated.)

Lens CoatingsAn important application of thin-film interference is in the coating of glass to make it “ nonreflecting,” particularly for lenses. A glass surface reflects about 4% of the light incident upon it. Good-quality cameras, microscopes, and other optical devices may contain six to ten thin lenses. Reflection from all these surfaces can reduce the light level considerably, and multiple reflections produce a background haze that reduces the quality of the image. By reducing reflection, transmission is increased. A very thin coating on the lens surfaces can reduce reflections considerably. The thickness of the film is chosen so that light (at least for one wavelength) reflecting from the front and rear surfaces of the film destructively interferes. The amount of reflection at a boundary depends on the difference in index of refraction between the two materials. Ideally, the coating material should have an index of refraction which is the geometric mean (= V n1 n2) of those for air and glass, so that the amount of reflection at each surface is about equal. Then destructive interference can occur nearly completely for one particular wavelength depending on the thickness of the coating. Nearby wavelengths w ill at least partially destructively interfere, but a single coating cannot eliminate reflections for all wavelengths. Nonetheless, a single coating can reduce total reflection from 4% to 1% of the incident light. Often the coating is designed to eliminate the center of the reflected spectrum (around 550 nm). The extremes of the spectrum— red and violet— w ill not be reduced as much. Since a mixture of red and violet produces purple, the light seen reflected from such coated lenses is purple (Fig. 34-22). Lenses containing two or three separate coatings can more effectively reduce a wider range of reflecting wavelengths.

*

c, O L V /

( ^ P H Y S I C S A P P L I E DLens coatings

FIGURE 34-22 A coated lens. Note color of light reflected from the front lens surface.

Q

Interference1. Interference effects depend on the simultaneous arrival

of two or more waves at the same point in space.2. Constructive interference occurs when the waves arrive

in phase with each other: a crest of one wave arrives at the same time as a crest of the other wave. The ampli­tudes of the waves then add to form a larger amplitude. Constructive interference also occurs when the path difference is exactly one full wavelength or any integer multiple of a full wavelength: 1A, 2A, 3A, • • •.

3. Destructive interference occurs when a crest of one wave arrives at the same time as a trough of the

other wave. The amplitudes add, but they are of opposite sign, so the total amplitude is reduced to zero if the two amplitudes are equal. Destructive interference occurs whenever the phase difference is half a wave cycle, or the path difference is a half-integral number of wavelengths. Thus, the total amplitude w ill be zero if two identical waves arrive one-half wavelength out of phase, or (m + |)A out of phase, where m is an integer.

4. For thin-film interference, an extra half-wavelength phase shift occurs when light reflects from an optically more dense medium (going from a material of lesser toward greater index of refraction).

SECTION 34-5 Interference in Thin Films 913

Coating

FIGURE 34-23 Example 34-8. Incident ray of light is partially reflected at the front surface of a lens coating (ray 1) and again partially reflected at the rear surface of the coating (ray 2), with most of the energy passing as the transmitted ray into the

FIGURE 34-24interferometer.

Michelson

EXAMPLE 34-8 Nonreflective coating. What is the thickness of an optical coating of MgF2 whose index of refraction is n = 1.38 and which is designed to eliminate reflected light at wavelengths (in air) around 550 nm when incident normally on glass for which n = 1.50?APPROACH We explicitly follow the procedure outlined in the Problem Solving Strategy on page 913.SOLUTION1. Interference effects. Consider two rays reflected from the front and rear

surfaces of the coating on the lens as shown in Fig. 34-23. The rays are drawn not quite perpendicular to the lens so we can see each of them. These two reflected rays w ill interfere with each other.

2. Constructive interference. We want to eliminate reflected light, so we do not consider constructive interference.

3. Destructive interference. To eliminate reflection, we want reflected rays 1 and 2 to be \ cycle out of phase with each other so that they destructively interfere. The phase difference is due to the path difference 21 traveled by ray 2, as well as any phase change in either ray due to reflection.

4. Reflection phase shift. Rays 1 and 2 both undergo a change of phase by \ cycle when they reflect from the coating’s front and rear surfaces, respectively (at both surfaces the index of refraction increases). Thus there is no net change in phase due to the reflections. The net phase difference w ill be due to the extra path 21 taken by ray 2 in the coating, where n = 1.38. We want 21 to equal \ A„ so that destructive interference occurs, where An = k /n is the wavelength in the coating. With 2 1 = An/ 2 = A /2 n, then

(550 nm)t = ^ = A =4 4 n

= 99.6 nm.(4)(1.38)

NOTE We could have set 21 = (m + ^)A„, where ra is an integer. The smallest thickness (ra = 0 ) is usually chosen because destructive interference w ill occur over the widest angle.NOTE Complete destructive interference occurs only for the given wavelength of visible light. Longer and shorter wavelengths w ill have only partial cancellation.

34—6 Michelson InterferometerA useful instrument involving wave interference is the Michelson interferometer (Fig. 34-24),f invented by the American Albert A. Michelson (Section 31-7). Monochromatic light from a single point on an extended source is shown striking a half-silvered mirror M s. This beam splitter mirror M s has a thin layer of silver that reflects only half the light that hits it, so that half of the beam passes through to a fixed mirror M 2, where it is reflected back. The other half is reflected by M s to a mirror M x that is movable (by a fine-thread screw), where it is also reflected back. Upon its return, part of beam 1 passes through M s and reaches the eye; and part of beam 2 , on its return, is reflected by M s into the eye. If the two path lengths are identical, the two coherent beams entering the eye constructively interfere and brightness w ill be seen. If the movable mirror is moved a distance A/4, one beam w ill travel an extra distance equal to A/2 (because it travels back and forth over the distance A/4). In this case, the two beams w ill destructively interfere and darkness w ill be seen. As M l is moved farther, brightness w ill recur (when the path difference is A), then darkness, and so on.

Very precise length measurements can be made with an interferometer. The motion of m irror M x by only ^A produces a clear difference between brightness and darkness. For A = 400 nm, this means a precision of 100 nm or 10-4mm! If m irror M x is tilted very slightly, the bright or dark spots are seen instead as a series of bright and dark lines or “ fringes.” By counting the number of fringes (or frac­tions thereof) that pass a reference line, extremely precise length measurements can be made.trThere are other types of interferometer, but Michelson’s is the best known.

914 CHAPTER 34 The Wave Nature of Light; Interference

34—7 Lum inous IntensityThe intensity of light, as for any electromagnetic wave, is measured by the Poynting vector in W /m 2, and the total power output of a source can be measured in watts (the radiant flux). But fo r measuring the visual sensation we call brightness, we must consider only the visible spectrum as well as the eye’s sensitivity to different wavelengths— the eye is most sensitive in the central, 550-nm (green), portion of the spectrum.

These factors are taken into account in the quantity luminous flux, Ft , whose unit is the lumen (lm ). One lumen is equivalent to ^ watts of 555-nm light.

Since the luminous flux from a source may not be uniform over all directions, we define the luminous intensity I£ as the luminous flux per unit solid anglef (steradian). Its unit is the candela (cd) where 1 cd = 1 lm /sr, and it is one of the seven basic quantities in the SI. (See Section 1-4 and Table 1-5.)

The illuminance, Et , is the luminous flux incident on a surface per unit area of the surface: = F J A . Its unit is the lumen per square meter (lm /m 2) and is a measure of the illum ination falling on a surface.*

EXAMPLE 34-9 Lightbulb illuminance. The brightness of a particular type of 100-W lightbulb is rated at 1700 lm. Determine (a) the luminous intensity and(b) the illuminance at a distance of 2.0 m.

APPROACH Assume the light output is uniform in all directions.SOLUTION (a) A fu ll sphere corresponds to 477 sr. Hence, It = 17001m/47r sr = 135 cd. It does not depend on distance. (b ) A t d = 2.0 m from the source, the luminous flux per unit area is

Ft 17001m ,

= (4* )(2 .0m )’ = 3 4 Im /m '

The illuminance decreases as the square of the distance.

f A solid angle is a sort of two-dimensional angle and is measured in steradians. Think of a solid angle starting at a point and intercepting an area A A on a sphere of radius r surrounding that point. The solid angle has magnitude A A/r2 steradians. A solid angle including all of space intercepts the full surface area of the sphere, 4irr2, and so has magnitude Airr2/r2 = Air steradians. (Compare to a normal angle, for which a full circle subtends 2tt radians.)*The British unit is the foot-candle, or lumen per square foot.

SummaryThe wave theory of light is strongly supported by the observa­tions that light exhibits interference and diffraction. Wave theory also explains the refraction of light and the fact that light travels more slowly in transparent solids and liquids than it does in air.

An aid to predicting wave behavior is Huygens’ principle, which states that every point on a wave front can be considered as a source of tiny wavelets that spread out in the forward direction at the speed of the wave itself. The new wave front is the envelope (the common tangent) of all the wavelets.

The wavelength of light in a medium with index of refraction n is

A„ = (34-1)

where A is the wavelength in vacuum; the frequency is not changed.

Young’s double-slit experiment clearly demonstrated the interference of light. The observed bright spots of the interference pattern are explained as constructive interference between the beams coming through the two slits, where the beams differ in path length by an integral number of wavelengths. The dark areas in between are due to destructive interference when the path lengths differ by |A ,|A , and so on. The angles 0 at which constructive interference occurs are given by

sin0 = ra (34-2a)d

where A is the wavelength of the light, d is the separation of the slits, and ra is an integer (0,1,2, •••). Destructive interference occurs at angles 0 given by

sin0 = (ra + (34-2b)

where ra is an integer (0,1,2, • • •).

Summary 915

The light intensity Ie at any point in a double-slit interference pattern can be calculated using a phasor diagram, which predicts that

Ie = I0 cos2 1 (34-6)

where Iq is the intensity at 6 = 0 and the phase angle 8 isO //8 = — sin 6. (34-4)

A

Two sources of light are perfectly coherent if the waves leaving them are of the same single frequency and maintain the

same phase relationship at all times. I f the light waves from the two sources have a random phase with respect to each other over time (as for two incandescent lightbulbs) the two sources are incoherent.

Light reflected from the front and rear surfaces of a thin film of transparent material can interfere constructively or destructively, depending on the path difference. A phase change of 180° or | A occurs when the light reflects at a surface where the index o f refraction increases. Such thin-film interference has many practical applications, such as lens coatings and using Newton’s rings to check the uniform ity of glass surfaces.

Questions1. Does Huygens’ principle apply to sound waves? To water

waves?2. What is the evidence that light is energy?3. Why is light sometimes described as rays and sometimes as

waves?4. We can hear sounds around corners but we cannot see

around corners; yet both sound and light are waves. Explain the difference.

5. Can the wavelength of light be determined from reflection or refraction measurements?

6 . Two rays o f light from the same source destructively inter­fere if their path lengths differ by how much?

7. Monochromatic red light is incident on a double slit and the interference pattern is viewed on a screen some distance away. Explain how the fringe pattern would change if the red light source is replaced by a blue light source.

8 . I f Young’s double-slit experiment were submerged in water, how would the fringe pattern be changed?

9. Compare a double-slit experiment for sound waves to that for light waves. Discuss the similarities and differences.

10. Suppose white light falls on the two slits o f Fig. 34-7, but one slit is covered by a red filte r (700 nm) and the other by a blue filte r (450 nm). Describe the pattern on the screen.

11. Why doesn’t the light from the two headlights o f a distant car produce an interference pattern?

12. Why are interference fringes noticeable only for a thin film like a soap bubble and not for a thick piece o f glass, say?

13. Why are Newton’s rings (Fig. 34-18) closer together farther from the center?

14. Some coated lenses appear greenish yellow when seen by reflected light. What wavelengths do you suppose the coating is designed to eliminate completely?

15. A drop of o il on a pond appears bright at its edges where its thickness is much less than the wavelengths of visible light. What can you say about the index of refraction of the o il compared to that o f water?

| Problems__________________34-2 Huygens' Principle1. ( II) Derive the law of reflection— namely, that the angle of

incidence equals the angle of reflection from a fla t surface— using Huygens’ principle for waves.

34-3 Double-Slit Interference2. (I) Monochromatic light falling on two slits 0.018 mm apart

produces the fifth-order bright fringe at a 9.8° angle. What is the wavelength of the light used?

3. (I) The third-order bright fringe o f 610 nm light is observed at an angle of 28° when the light falls on two narrow slits. How far apart are the slits?

4. ( II) Monochromatic light falls on two very narrow slits 0.048 mm apart. Successive fringes on a screen 6.00 m away are 8.5 cm apart near the center of the pattern. Determine the wavelength and frequency o f the light.

5. ( II) I f 720-nm and 660-nm light passes through two slits 0.68 mm apart, how far apart are the second-order fringes fo r these two wavelengths on a screen 1.0 m away?

6 . ( II) A red laser from the physics lab is marked as producing 632.8-nm light. When light from this laser falls on two closely spaced slits, an interference pattern formed on a wall several meters away has bright fringes spaced 5.00 mm apart near the center o f the pattern. When the laser is replaced by a small laser pointer, the fringes are 5.14 mm apart. What is the wavelength of light produced by the pointer?

7. (II) Light of wavelength A passes through a pair o f slits separated by 0.17 mm, forming a double-slit interference pattern on a screen located a distance 35 cm away. Suppose that the image in Fig. 34-9a is an actual-size reproduction of this interference pattern. Use a ruler to measure a pertinent distance on this image; then utilize this measured value to determine A (nm).

8 . (II) Light o f wavelength 680 nm falls on two slits and produces an interference pattern in which the third-order bright fringe is 38 mm from the central fringe on a screen2.6 m away. What is the separation of the two slits?

9. (II) A parallel beam of light from a H e-N e laser, w ith a wavelength 633 nm, falls on two very narrow slits 0.068 mm apart. How far apart are the fringes in the center o f the pattern on a screen 3.8 m away?

10. (II) A physics professor wants to perform a lecture demon­stration o f Young’s double-slit experiment for her class using the 633-nm light from a He-N e laser. Because the lecture hall is very large, the interference pattern w ill be projected on a wall that is 5.0 m from the slits. For easy viewing by all students in the class, the professor wants the distance between the m = 0 and m = 1 maxima to be25 cm. What slit separation is required in order to produce the desired interference pattern?

916 CHAPTER 34 The Wave Nature of Light; Interference

11. ( II) Suppose a thin piece of glass is placed in front o f the 3 4 lower slit in Fig. 34-7 so that the two waves enter the slits 23. 180° out o f phase (Fig. 34-25). Describe in detail the inter­ference pattern on the screen.

w w m

FIGURE 34-25Problem 11.

12. ( II) In a double-slit experiment it is found that blue light of wavelength 480 nm gives a second-order maximum at a 26. certain location on the screen. What wavelength of visiblelight would have a minimum at the same location?

13. ( II) Two narrow slits separated by 1.0 mm are illuminatedby 544 nm light. Find the distance between adjacent bright 27. fringes on a screen 5.0 m from the slits.

14. ( II) In a double-slit experiment, the third-order maximum for light o f wavelength 500 nm is located 12 mm from the central bright spot on a screen 1.6 m from the slits. Light of wavelength 650 nm is then projected through the same slits.How far from the central bright spot w ill the second-order maximum of this light be located?

15. ( II) Light o f wavelength 470 nm in air falls on two slits6.00 X 10 2 mm apart. The slits are immersed in water, as is a viewing screen 50.0 cm away. How far apart are the fringes on the screen?

16. ( II) A very thin sheet o f plastic (n = 1.60) covers one slit of a double-slit apparatus illuminated by 680-nm light. The center point on the screen, instead o f being a maximum, is dark. What is the (minimum) thickness of the plastic?

34-4 Intensity in Two-Slit Interference2oa

17. (I) I f one slit in Fig. 34-12 is covered, by what factor does the intensity at the center of the screen change?

18. ( II) Derive an expression similar to Eq. 34-2 which gives the angles for which the double-slit intensity is one-half its maximum value, I6 = \Iq . 29

19. ( II) Show that the angular fu ll w idth at half maximum of the central peak in a double-slit interference pattern is given by A 6 = A/2d if A « d.

20. (II) In a two-slit interference experiment, the path length to a certain point P on the screen differs for one slit in compar- 30. ison with the other by 1.25A. (a) What is the phase difference between the two waves arriving at point P? (b ) Determinethe intensity at P, expressed as a fraction of the maximum intensity I0 on the screen.

21. ( I ll) Suppose that one slit o f a double-slit apparatus is 3 1 . wider than the other so that the intensity o f light passing through it is twice as great. Determine the intensity I as a function of position (0 ) on the screen for coherent light.

22. ( I ll) (a) Consider three equally spaced and equal-intensity 32. coherent sources of light (such as adding a th ird slit to thetwo slits o f Fig. 34-12). Use the phasor method to obtain the intensity as a function of the phase difference 8 (Eq. 34-4). (b ) Determine the positions of maxima and minima.

-5 Thin-Film Interference(I) I f a soap bubble is 120 nm thick, what wavelength is most strongly reflected at the center o f the outer surface when illuminated normally by white light? Assume that n = 1.32.

(I) How far apart are the dark fringes in Example 34-6 if the glass plates are each 28.5 cm long?

(II) (a) What is the smallest thickness of a soap film (n = 1.33) that would appear black if illuminated w ith 480-nm light? Assume there is air on both sides of the soap film , (b) What are two other possible thicknesses for the film to appear black? (c) I f the thickness t was much less than A, why would the film also appear black?

(II) A lens appears greenish yellow ( A = 570 nm is strongest) when white light reflects from it. What minimum thickness of coating (n = 1.25) do you think is used on such a glass (n = 1.52) lens, and why?

(II) A thin film of o il (nQ = 1.50) w ith varying thickness floats on water (nw = 1.33). When it is illuminated from above by white light, the reflected colors are as shown in Fig. 34-26. In air, the wavelength o f yellow light is 580 nm.(a) Why are there no reflected colors at point A? (b) What is the o il’s thickness t at point B?

FIGURE 34-26Problem 27.

(II) A thin o il slick (nQ = 1.50) floats on water (nw = 1.33). When a beam of white light strikes this film at normal incidence from air, the only enhanced reflected colors are red (650 nm) and violet (390 nm). From this in for­mation, deduce the (minimum) thickness t of the o il slick.

(II) A total of 31 bright and 31 dark Newton’s rings (not counting the dark spot at the center) are observed when 560-nm light falls normally on a planoconvex lens resting on a fla t glass surface (Fig. 34-18). How much thicker is the center than the edges?

(II) A fine metal fo il separates one end o f two pieces of optically fla t glass, as in Fig. 34-20. When light of wavelength 670 nm is incident normally, 28 dark lines are observed (w ith one at each end). How thick is the foil?

(II) How thick (minimum) should the air layer be between two fla t glass surfaces if the glass is to appear bright when 450-nm light is incident normally? What if the glass is to appear dark?

(II) A uniform thin film o f alcohol (n = 1.36) lies on a fla t glass plate (n = 1.56). When monochromatic light, whose wavelength can be changed, is incident normally, the reflected light is a minimum for A = 512 nm and a maximum for A = 635 nm. What is the minimum thickness of the film?

I I.:. k

Water

Problems 917

33. (II) Show that the radius r o f the mth dark Newton’s ring, as viewed from directly above (Fig. 34-18), is given by r = V'm XR where R is the radius of curvature of the curved glass surface and A is the wavelength of light used. Assume that the thickness of the air gap is much less than R at all points and that r <SC R. [Hint: Use the binomial expansion.]

34. (II) Use the result o f Problem 33 to show that the distance between adjacent dark Newton’s rings is

Ar

fo r the mth ring, assuming ra » 1 .

35. ( II) When a Newton’s ring apparatus (Fig. 34-18) is immersed in a liquid, the diameter o f the eighth dark ring decreases from 2.92 cm to 2.54 cm. What is the refractive index o f the liquid? [Hint: see Problem 33.]

36. ( II) A planoconvex lucite lens 3.4 cm in diameter is placed on a fla t piece o f glass as in Fig. 34-18. When 580-nm light is incident normally, 44 bright rings are observed, the last one right at the edge. What is the radius o f curvature o f the lens surface, and the focal length of the lens? [Hint: see Problem 33.]

37. ( II) Let’s explore why only “ th in ” layers exhibit th in-film interference. Assume a layer o f water, sitting atop a fla t glass surface, is illum inated from the air above by white ligh t (a ll wavelengths from 400 nm to 700 nm). Further, assume that the water layer’s thickness t is much greater than a micron (= 1 0 0 0 nm); in particular, le t t = 200 ftm. Take the index of refraction for water to be n = 1.33 fo r all visible wavelengths, (a) Show that a visible color w ill be reflected from the water layer i f its wavelength is A = 2nt/m, where ra is an integer. (b) Show that the two extremes in wavelengths (400 nm and 700 nm) of the incident ligh t are both reflected from the water layer and determine the m-value associated w ith each, (c) How many other visible wavelengths, besides A = 400 nm and 700 nm, are reflected from the “ th ick” layer o f water? (d ) How does this explain why such a thick layer does not reflect colorfully, but is white or grey?

38. ( I ll) A single optical coating reduces reflection to zero fo r A = 550 nm. By what factor is the intensity reduced by the coating for A = 430 nm and A = 670 nm as compared to no coating? Assume normal incidence.

| General Problems___________

34-6 Michelson Interferometer“ 39. (II) How far must the m irror M i in a Michelson interferometer

be moved if 650 fringes o f 589-nm light are to pass by a reference line?

15 40. (II) What is the wavelength of the light entering an interferometer if 384 bright fringes are counted when the movable m irror moves 0.125 mm?

“ 41. (II) A micrometer is connected to the movable m irror o f an interferometer. When the micrometer is tightened down on a thin metal fo il, the net number of bright fringes that move, compared to the empty micrometer, is 272. What is the thickness o f the foil? The wavelength o f light used is 589 nm.

“42. ( I ll) One o f the beams of an interferometer (Fig. 34-27) passes through a small evacuated glass container 1.155 cm deep. When a gas is allowed to slowly f ill the container, a total of 176 dark fringes are counted to move past a refer­ence line. The light used has a wavelength o f 632.8 nm. Calculate the index o f refraction of the gas at its final density, assuming that the interferometer is in vacuum.

To mirror M

Source -

M,7

Glass container

h1.155 cmMo

FIGURE 34-27 Problem 42.

“43. ( I ll) The yellow sodium D lines have wavelengths of 589.0 and 589.6 nm. When they are used to illum inate a Michelson interferometer, it is noted that the interference fringes disappear and reappear periodically as the m irror M i is moved. Why does this happen? How far must the m irror move between one disappearance and the next?

“34-7 Luminous Intensity“44. (II) The illuminance o f direct sunlight on Earth is about

105 lm /m 2. Estimate the luminous flux and luminous intensity of the Sun.

“45. (II) The luminous efficiency o f a lightbulb is the ratio of luminous flux to electric power input, (a) What is the luminous efficiency (%) o f a 100-W, 1700-lm bulb? (b ) How many 40-W, 60-lm /W fluorescent lamps would be needed to provide an illuminance o f 250 lm /m 2 on a factory floor of area 25 m X 30 m? Assume the lights are 10 m above the floor and that half their flux reaches the floor.

46. Light of wavelength 5.0 X 10_7m passes through two parallel slits and falls on a screen 4.0 m away. Adjacent bright bands of the interference pattern are 2.0 cm apart.(a) Find the distance between the slits, (b) The same two slits are next illuminated by light o f a different wavelength, and the fifth-order minimum for this light occurs at the same point on the screen as the fourth-order minimum for the previous light. What is the wavelength o f the second source o f light?

47. Television and radio waves reflecting from mountains or airplanes can interfere w ith the direct signal from the station, (a) What kind of interference w ill occur when 75-MHz television signals arrive at a receiver directly from a distant station, and are reflected from a nearby airplane 122 m directly above the receiver? Assume \ \ change in phase of the signal upon reflection, (b) What kind of interference w ill occur if the plane is 22 m closer to the receiver?

918 CHAPTER 34 The Wave Nature of Light; Interference

48. A radio station operating at 88.5 MHz broadcasts from two identical antennas at the same elevation but separated by a9.0-m horizontal distance d, Fig. 34-28. A maximum signal is found along the midline, perpendicular to d at its m idpoint and extending horizontally in both directions. I f the m idline is taken as 0°, at what other angle(s) 6 is a maximum signal detected? A minimum signal? Assume all measurements are made much farther than 9.0 m from the antenna towers.

d - Midline

FIGURE 34-28Problem 48. Antenna 2

55.

56.

Light o f wavelength A strikes a screen containing two slits a distance d apart at an angle 6\ to the normal. Determine the angle 6m at which the mth-order maximum occurs.

Consider two antennas radiating 6.0-MHz radio waves in phase with each other. They are located at points S1 and S2, separated by a distance d = 175 m, Fig. 34-30. Determine the points on the y axis where the signals from the two sources w ill be out o f phase (crests of one meet troughs of the other).

FIGURE 34-30Problem 56.

S2

175 m

49. Light o f wavelength 690 nm passes through two narrow slits 0.66 mm apart. The screen is 1.60 m away. A second source of unknown wavelength produces its second-order fringe 1.23 mm closer to the central maximum than the 690-nm light. What is the wavelength o f the unknown light?

50. Monochromatic light of variable wavelength is incident normally on a thin sheet o f plastic film in air. The reflected light is a minimum only for A = 491.4 nm and A = 688.0 nm in the visible spectrum. What is the thickness of the film (n = 1.58)? [H int Assume successive values of ra.]

15 51. Suppose the m irrors in a Michelson interferometer are perfectly aligned and the path lengths to m irrors M i and M 2 are identical. W ith these in itia l conditions, an observer sees a bright maximum at the center of the viewing area. Now one of the m irrors is moved a distance x. Determine a formula for the intensity at the center o f the viewing area as a function o f x, the distance the movable m irror is moved from the in itia l position.

52. A highly reflective m irror can be made for a particular wavelength at normal incidence by using two thin layers of transparent materials of indices o f refraction ni and n2 ( l < < n2) on the surface of the glass (n > n2). What should be the minimum thicknesses d \ and d2 in Fig. 34-29 in terms of the incident wavelength A, to maximize reflection?

57. What is the minimum (non-zero) thickness for the air layer between two fla t glass surfaces if the glass is to appear dark when 680-nm light is incident normally? What if the glass is to appear bright?

58. L lo yd ’s mirror provides one way o f obtaining a double-slit interference pattern from a single source so the light is coherent. As shown in Fig. 34-31, the light that reflects from the plane m irror appears to come from the virtual image of the slit. Describe in detail the interference pattern on the screen.

FIGURE 34-31 Problem 58.

nj n2di d2FIGURE 34-29

Problem 52.

53. Calculate the minimum thickness needed for an antireflec- tive coating (n = 1.38) applied to a glass lens in order to eliminate (a) blue (450 nm), or (b) red (720 nm) reflections for light at normal incidence.

54. Stealth aircraft are designed to not reflect radar, whose wavelength is typically 2 cm, by using an antireflecting coating. Ignoring any change in wavelength in the coating, estimate its thickness.

59. Consider the antenna array of Example 34-5, Fig. 34-15. Let d = A/2, and suppose that the two antennas are now 180° out o f phase w ith each other. Find the directions for constructive and destructive interference, and compare with the case when the sources are in phase. (These results illus­trate the basis for directional antennas.)

60. Suppose you viewed the light transmitted through a thin film layered on a fla t piece of glass. Draw a diagram, similar to Fig. 34-17 or 34-23, and describe the conditions required for maxima and minima. Consider all possible values of index of refraction. Discuss the relative size of the minima compared to the maxima and to zero.

61. A thin film of soap (n = 1.34) coats a piece o f fla t glass (n = 1.52). How thick is the film if it reflects 643-nm red light most strongly when illuminated normally by white light?

Screen^

Source

° \ MirrorVirtual image of slit

General Problems 919

62. Two identical sources Si and S2, separated by distance d, coherently emit light of wavelength A uniformly in all direc­tions. Defining the x axis with its origin at Si as shown in Fig. 34-32, find the locations (expressed as multiples of A) where the signals fromthe two sources are out Sx ______________________xof phase along this axis for x > 0, if d = 3A.

d

FIGURE 34-32Problem 62. S2u

63. A two-slit interference set-up with slit separation d = 0.10 mm produces interference fringes at a particular set of angles 0m (where m = 0 , 1, 2 ,...) for red light of frequency / = 4.6 X 1014 Hz. I f one wishes to construct an analogous two-slit interference set-up that produces interference fringes at the same set of angles 6m for room-temperature sound of middle-C frequency / s = 262 Hz, what should the slit sepa­ration ds be for this analogous set-up?

64. A radio telescope, whose two antennas are separated by 55 m, is designed to receive 3.0-MHz radio waves produced by astronomical objects. The received radio waves create3.0-MHz electronic signals in the telescope’s left and right antennas. These signals then travel by equal-length cables to a centrally located amplifier, where they are added together. The telescope can be “pointed” to a certain region of the sky by adding the instantaneous signal from the right antenna to a “ time-delayed” signal received by the left antenna a time At ago. (This time delay of the left signal can be easily accomplished with the proper electronic circuit.) If a radio astronomer wishes to “view” radio signals arriving from an object oriented at a 12° angle to the vertical as in Fig. 34-33, what time delay A? is necessary?

65. In a compact disc (CD), digital information is stored as a sequence of raised surfaces called “pits” and recessed surfaces called “ lands.” Both pits and lands are highly reflec­tive and are embedded in a thick plastic material with index of refraction n = 1.55 (Fig. 34-34). As a 780-nm wave­length (in air) laser scans across the pit-land sequence, the transition between a neighboring pit and land is sensed by monitoring the intensity of reflected laser light from the CD. A t the moment when half the width of the laser beam is reflected from the pit and the other half from the land, we want the two reflected halves of the beam to be 180° out of phase with each other. What should be the (minimum) height difference t between a pit and land? [When this light enters a detector, cancellation of the two out-of-phase halves of the beam produces a minimum detector output.]

ktrlkyjiiiti______•r Ij id"

i 1— I F -

Plastici

tt - 1.55 iiiitipiit1\I

FIGURE 34-34 ^lserProblem 65. beam

* Numerical/Computer* 66. (II) A Michelson interferometer can be used to determine the

index of refraction of a glass plate. A glass plate (thickness t) is placed on a platform that can rotate. The plate is placed in the light’s path between the beam splitter and either the fixed or movable mirror, so that its thickness is in the direc­tion of the laser beam. The platform is rotated to various angles, and the number of fringes shifted is counted. It can be shown that if N is the number of fringes shifted when the angle of rotation changes by 0 , the index of refraction is n = (21 - NX)(1 - cos0)/[2£(l - cos0) - NX] where t is the thickness of the plate. The accompanying Table shows the data collected by a student in determining the index of refraction of a transparent plate by a Michelson interferometer.

N 25 50 75 100 125 1500 (degree) 5.5 6.9 8.6 10.0 11.3 12.5

In the experiment A = 632.8 nm and t = 4.0 mm. Deter­mine n for each 0 and find the average n.

Answers to Exercises B: (a) If). (b) 0.5o /0; (c) o.

A: 333 nm; 6.0 X 1014Hz;2.0 X 108 m/s. C: (e).

920 CHAPTER 34 The Wave Nature of Light; Interference