Solution For Optics Pedrotti

Chapter 1 Nature of Light
h h 6.63 × 10−34 J · s
1-1. a) λ = = = = 6.63 × 10−34 m
p m v (0.05 kg) (20 m/s)
h h 6.63 × 10−34 J · s
b) λ = =√ = = 3.88 × 10−10 m
p 2 m E [(2 · 9.11 × 10−31 kg) (10 · 1.602 × 10−19 J)]1/2
Energy n h ν n h c 100 6.63 × 10−34 J · s 3 × 108m/s

1-2. P = = = = = 3.62 × 10−17 W
time t tλ (1 s) (550 × 10−9m)
1-3. The energy of a photon is given by E = h ν = h c/λ
6.63 × 10−34 J · s 3 × 108m/s

1 eV
At λ = 380 nm: E = = 5.23 × 10−19 J = 3.27 eV

380 × 10−9 m 1.60 × 10−19 J
6.63 × 10 −34
J · s 3 × 10 m/s
8

1 eV
At λ = 770 nm: E = = 2.58 × 10−19 J = 1.61 eV

770 × 10 m−9 1.60 × 10−19 J
h hc hc h
1-4. p = E/c = m c2/c = m c = 2.73 × 10−22 kg · m/s, λ= = = = = 2.43 × 10−12 m
p E m c2 m c
2 1 MeV
1-5. Ev=0 = m c2 = 9.109 × 10−31 kg 2.998 × 108 m/s = 8.187 × 10−14 J = .511 MeV

1.602 × 10−19 J
√ √
1-6. c p = E 2 − m2 c4 , where E = EK + m c2 = (1 + 0.511))MeV. So c p = 1.5112 − 0.5112 MeV
That is, c p = 1.422 MeV and p = 1.422 MeV/c.
6.626 × 10−34 J · s 2.998 × 108 m/s

1 eV 1Å 12, 400

hc
Å · eV

1-7. λ = = =
E E 1.602 × 10−19 J 10−10 m E
!
1 h −1/2 i 1
1-8. EK = m c2 − 1 = m c2 1 − v 2/c2 − 1 ≃ m c2 1 − ( − 1/2) v 2/c2 − 1 = m v 2

2
p
1 − v 2/c2
1
1-9. The total energy of the proton is,
9 1.60 × 10 J
−19

2
E = EK + m p c = 2 × 10
2
+ 1.67 × 10−27 kg 3.00 × 108 m/s = 4.71 × 10−10 J

1 eV
q h 2 2 2 i−1/2
E 2 − m2p c4 4.71 × 10−10 J − 1.67 × 10−27 kg 3.00 × 108 m
a) p = =
c 3.00 × 108 m/s
p = 1.49 × 10−18 kg · m/s
b) λ = h/p = 6.63 × 10−34 J · s / 1.49 × 10−18 kg · m/s = 4.45 × 10−16 m

c) λphoton = h c/E = 6.63 × 10−34 J · s 3.00 × 108 m/s / 4.71 × 10−10 = 4.22 × 10−16 m

1000 W/m2 10−4 m2

Energy Energy
1-10. nphotons = = = = 2.77 × 1017
hν h c/λ (6.63 × 10−34 J) (3.00 × 108 m/s)/(550 × 10−9 m)
n1 Ee/h ν1 Ee λ1/h c λ1
1-11. = = =
n2 Ee/h ν2 Ee λ2/h c λ2
1-12. The wavelength range is 380 nm to 770 nm. The corresponding frequencies are
c 3.00 × 108 m/s c 3.00 × 108 m/s

ν770 = = = 3.89 × 1014 Hz ν380 = = = 7.89 × 1014 Hz
λ 770 × 10 m
−9 λ 380 × 10−9 m
1-13. The wavelength of the radio waves is λ = c/ν = 3.00 × 108 m/s / 100 × 106 Hz = 3 m. The length of the

half wave antenna is then λ/2 = 1.5 m.
1-14. The wavelength is λ = c/ν = 3.0 × 108 m/s / 90 × 106 Hz = 3.33 m. The length of each of the rods is then

λ/4 = 0.83 m.
1-15. a) t = Dl/c = (90 × 103 /3.0 × 108 s = 3.0 × 10−4 s. b) Ds = vs t = (340 ) 3.0 × 10−4 m = 0.10 m

Φe 500 W Φ 500 W
1-16. a) Ie = = = 39.8 W/sr b) Me = e = = 106 W/m2
∆ω 4 π sr A 5 × 10−4 m2
Φ Φe 500 W
c) Ee = e = = 9.95 W/m2 e) Φe = Ee A = 9.95 W/m2 π (0.025 m)2 = .0195 W

2
= 2
A 4πr 4 π(2 m)
1-17. a) The half angle divergence θ1/2 can be found from the relation
rsp ot 0.0025 m
tan(θ1/2) ≈ θ1/2 = = = 1.67 × 10−4 rad = .0096◦
Lro om 15 m
2 2
Aspot π rspot π (0.0025 m)
b) The solid angle is ∆ω = 2 = 2 = 2 = 8.73 × 10−8 sr.
Lro om Lro om (15 m)
Φe Φ 0.0015 W
c) The irradiance on the wall is Ee = = 2e = = 76.4 W/m2.
Aspot π rsp ot π (0.0025 m)2
d) The radiance is (approximating differentials as increments)
Φe 0.0015 W W
Le ≈ = = 8.75 × 1010 2
∆ω ∆Alaser cosθ (8.73 × 10−8 sr) π (0.00025 m)2 cos(0) m · sr

2
Chapter 2 Geometrical Optics
P P
dop i ni xi
2-1. t = =
c c
2-2. Referring to Figure 2 12 and with lengths in cm,
1/2 1/2
n0 x2 + y 2 + ni y 2 + so + si − x)2 = n o so + n i si
2 1/2
1/2
(1) x2 + y 2 + 1.5 y 2 + (30 − x) = 20 + 1.5 (10) = 35
1/2 2
2.25 y 2 + (30 − x)2 = 35 − x2 + y 2
1/2
1.25 x2 + y2 + 70 x2 + y 2

− 135 x + 800 = 0
Using a calculator to guess and check or using a computer algebra system, (like the free program Maxima,
for example) one can numerically solve this equation for x for given y values. Doing so results in,
x (cm) 20 20.2 20.4 20.8 21.6 22.4 23.2 24.0 24.8 25.6 26.4 27.2
y (cm) 0 ± 1.0 ± 1.40 ± 1.96 ± 2.69 ± 3.20 ± 3.58 ± 3.85 ± 4.04 ± 4.14 ± 4.18 ± 4.13
2-3. Refer to the figure for the relevant parameters.
√
d = d ′ = 302 + 2.52 = 30.104cm d
t
d′
Fermat: d + d ′ = s + s ′ − t + m t
d + d ′ = s + s ′ + t (m − 1)
2 (30.10399) = 60 = t (1.52 − 1)
t = 4mm
s ′ = 30 cm s ′ = 30 cm
n = 1.52
2-4. See the figure below. Let the height of the person be h = h1 + h2.
h1 The person must be able to see the top of

2
h1
his head and the bottom of his feet. From
the figure it is evident that the mirror
height is:
mirror hm irror = h − h1/2 − h2/2 = h/2

h2
h2 The mirror must be half the height of the

2 person. So for a person of height six ft
person, the mirror must be 3 ft high.
2-5. Refer to the figure.
45 ◦
Top
√
At Top: (1) sin 45 = 2 sin θ ′ ⇒ θ ′ = 30
30 ◦ √ √
At Side: 2 sin 60◦ = (1) sin θ ′, sin θ ′ = 1.5 > 1
Side Thus total internal reflection occurs.
60 ◦
At Bottom: reverse of Top: θ ′ = 45◦
45 ◦
Bottom
3
2-6. The microscope first focuses on the scratch using direct rays. Then it focuses on the image I2 formed in a
two step process: (1) reflection from the bottom to produce an intermediate image I1 and (2) refraction
through the top surface to produce an image I2. Thus, I1 is at 2t from top surface, and I2 is at the
2t 2t 3
apparent depth for I1, serving as the object: s ′ = or n = ′ = = 1.60
n s 1.87
7.60/4
2-7. Refer to Figure 2 33 in the text. By geometry, tan θc = so θc = 40.18◦
1 2.25
Snell’s law: n sin θc = (1) sin 90◦ ⇒ n = = 1.55
sin 40.18◦
2-8. Referring to the figure one can see that,

t
s = AB sin (θ1 − θ2) and AB = . Therefore,
cos θ2
t sin (θ1 − θ2) θ1 A n1
s= . For t = 3 cm, n2 = 1.50, θ1 = 50◦, n2
cos θ2 n 1
Snell’s law gives, sin θ2 = 1 sin θ1 = sin 50◦. t θ1 − θ2
n2 1.5 θ2
3 sin (50◦ − 30.71◦)
Then, θ2 = 30.71◦ and s = = 1.153cm.
cos 30.71◦ B s
1 1 1
2-9. Image of near end: s = 60cm, = = , s ′ − 24cm
60 s ′ − 40
1 1 1
Image of far end: s = 60 + 100 cm, + = , s ′ = − 32cm. So, L ′ = ∆s ′ = − 24 − ( − 32) = 8cm
160 s ′ − 40
2-10. (a) See Figure 2 34 in the text. Image due to rays directly from bubble through plane interface:
n1 n2 1.5 1
+ ′ = 0 or + ′ = 0 ⇒ s ′ = − 3.33 cm.
s s s s
(b) Image due to rays first reflected in spherical mirror, then refracted through plane interface:
1 1 2 1 1 2
reflection: + =− and + =− s ′ = − 7.5cm
2 s1′ R 2.5 s1′ − 7.5 1
n n 1.5 1
refraction: 1 + ′2 = 0 or
′
+ = 0 s2 = − 10cm
s s2 15 s2′
Thus the images are at 3.33 cm and 10 cm behind the interface.
2-11. There are 5 unknowns: s1 and s1′ in position (1), s2 and s2′ in position (2), and the focal length f of the
mirror. The five equations that, solved simultaneously, yield the results are:
(1) linear magnification: s1′ /s1 = 2 (2) linear magnification: s2′ /s2 = 3
s s′ s s′
(3) focal length from mirror equation: f = 1 1 ′ (4) focal length from mirror equation: f = 2 2 ′
s1 + s1 s2 + s2
(5) image distance relation: s2′ = s1′ + 75
One finds s1 = 112.5 cm, s2 = 100 cm, s1′ = 225 cm, s2′ = 300 cm, f = 75cm
2-12. The object distance from the front surface is the diameter of the sphere, 5 cm. Then,
′
n1 n2 n2 − n1 1.5 1 1 − 1.5 n s (1.5) ( − 10)
⇒ s = − 10 cm and, m = − 1 = −
′
+ ′ = ⇒ + ′= = + 3.
s s R 5 s − 2.5 n2s (1) (5)
n1 n2 n2 − n1
2-13. Generally, + ′ =
s s R
n1 n2 n2 − n1 n R
(a) + = or f = 1 (b) n2 > n1: then R > 0 (convex), n2 < n1: then R < 0 (concave)
f ∞ R n2 − n1
4
n1 n2 n2 − n1 n s′
2-14. (a) In this position the object distance is s = 15 cm. Then, using,
+ ′= and m = − 1 ,
s s R n2 s
4 1 1 − 4/3 ′ (4/3) ( − 15)
+ = ⇒ s = − 15 cm (center) and, m = − ⇒ m = 4/3.
3 (15) s′ − 15 (1) (15)
(b) Similarly, in this position s = 7.5 cm so that,
4 1 1 − 4/3 n s′ (4/3) ( − 45/7)
+ ′= ⇒ s ′ = − 6.4 cm, m = − 1 = − = 8/7.
3 (15/2) s − 15 n2 s (1) (15/2)
2-15. See Figure 2 35 in the text. Rays from the object are (a) refracted through the spherical window, (b) then
reflected from the back plane mirror, (c) then refracted out again through the spherical window. Taking
these in turn:
n n n − n1 1 4 4/3 − 1 n s′ (1) (40)
(a) 1 + 2′ = 2 ⇒ + ′= ⇒ s ′ = 40 cm. Then, m = 1 = =−1
s s R 30 3s 5 n2 s (4/3) (30)
(b) s = 25 − 40 = − 15 cm (virtual object), s ′ = − s = 15 cm, m = − s ′/s = 1
4/3 1 1 − 4/3 − (4/3) ( − 15)
(c) + ′= ⇒ s ′ = − 15cm. Then, m = = + 2.
10 s −5 (1) (10)
The overall magnification is m = ( − 1) ( + 1) ( + 2) = − 2. Thus a virtual, inverted, double sized image
appears 15 cm behind (right) the spherical window.
2-16. The plane side of the lens has R1 = ∞. The radius of curvature R2 of the convex side is then found from
the lensmaker’s equation:

1 n2 − n1 1 1 1 1.52 − 1 1 1
= − ⇒ = − ⇒ R2 = − 13cm
f n1 R1 R2 25 1 ∞ R2

1 n − n1 1 1
2-17. In general the lensmaker’s equation gives, = 2 −
f n1 R1 R2
For the positive meniscus
shown to the right, R1 = 5 cm and R2 = 10 cm.
lens
1 1.50 − 1 1 1 R1
Then, = − ⇒ f = + 20 cm
f 1 5 10 R2
For the negative meniscus lens

shown to the right, R1 = 10 cm and R2 = 5 cm. R1
1 1.50 − 1 1 1
For this case, = − ⇒ f = − 20 cm
f 1 10 5 R2
2-18. The thin lens equation assumes identical, refractive indices on both sides. In this case we can modify the
procedure, beginning with Eq. (2 23), to allow for three distinct media as shown.
n1 n2 n2 − n1
20 cm + = , left lens surface
s1 s1′ R1
n2 n2 n3 n3 − n2
Fish Tank
+ = , right lens surface
n1
n3 s2 s2′ R2
n1 n3 n2 − n1 n3 − n2
For a thin lens, s2 ≈ − s1′ . Adding the equations, + = + . Or, simply,
s1 s2′ R1 R2
n1 n3 n2 − n1 n3 − n2 4 1 3/2 − 4/3 1 − 3/2
+ ′= + ⇒ + = + , so that s ′ = − 22.5cm.
s s R1 R2 3 (20) s ′ 30 − 30
n1 s1′ n2 s2′

The total magnification is mT = m1 m2 = − − , where s2 = − s1′ . So,
n2 s n 3 s2
n s′ (4/3) ( − 22.5)
mT = − 1 = − = 1.50.
n3 s (1) (20)
5
1 1 1 1 1 1
2-19. (a) Using = + as in Eq. (2 33), = + or feq = 6.67cm
feq f1 f2 feq − 5 20
(b) A pair of separated lenses has a front and a back focal length. The front focal length is the object posi
tion from the first lens that leads to an image at infinity. The back focal length is the image position for an
object at infinity. These cases are illustrated below. The drawings are generic and not to scale.
L L
feq, f feq,b
1 2 1 2
1 1 1
Working backwards, for the front focal length: Lens 2:
+ = or s2 = f2 , s2 = L − s1′ or s1′ = L − f2.
s2 ∞ f 2
1 1 1 f (L − f2) ( − 5) (10 − 20)
Lens 1: + = or feq,f = 1 = cm = − 10cm
feq,f L − f2 f1 L − (f1 + f2) 10 − ( − 5 + 20)
For the back focal length: for lens 1:
1 1 1 1 1 1
+ = ⇒ + = ⇒ s1′ = f1. Then s2 = L − f1, so that, for lens 2:
s1 s1′ f1 ∞ s1′ f1
1 1 1 1 1 1 f2 (L − f1) (20) (10 − ( − 5))
+ = ⇒ + = ⇒ feq,b = = cm = − 60 cm
s2 s2′ f2 L − f1 feq,b f2 (L − f1) − f2 (10 − ( − 5)) − 20
2-20. See Figure 2 36 in the text. Consider the three media as a sequence of three thin lenses. Each has a focal
length given by the lensmaker’s equation, and the equivalent focal length is given Eq. (2 33) as,

1 1 1 1 1 1 1
= + + . Then, = (1.5 − 1) − ⇒ f1 = 30cm,
feq f1 f2 f3 f1 ∞ − 15

1 1 1 150 1
= (1.65 − 1) − or f2 = − cm, and = same as for f1 : f3 = 30 cm. Then,
f2 − 15 15 13 f3
1 1 − 13 1
= + + and so feq = − 50cm.
feq 30 150 30
2-21. (a) One can use the formula derived in problem 2 19b, or do the calculation at first hand:
1 1 1 1 1 1
Second lens: + = or s2 = 20cm, First lens: + = or s1 = 3.33cm. The object should be
s2 ∞ 20 s1 − 4 20
placed 3.33 cm before the first lens.
(b) In the figure below the dashed arrow is the intermediate image that acts as the object for the second
lens. Since the image is “at infinity” it is described by an angular magnification. The image appears erect
and magnified.
F2
F1 F2 F1
To eye
1 2
6
2-22. Refer to Figure 2 37 in the text.
1 1 1 s′ − 3f /5
(b) Lens heading towards mirror: + ′= or s ′ = − 3 f /5 . m1 = − = − = 2/5
3 f /2 s −f s 3f /2
Mirror:
5 1 1 s′
s = 3 f + 3 f /5 = 18 f /5 ⇒ + ′ = ⇒ s ′ = 18 f /13, m2 = − = − (18 f /13)/(18 f /5) = − 5/13
18 f s f s
Lens after reflection:
13 1 1 13
s = 3 f − 18 f /13 = 21 f /13 ⇒ + = or s ′ = 21 f /34, m3 = − s ′/s = − (21 f /34)/(21 f /13) =
21 f s ′ − f 34

2 5 13
mT = − = − 17. The image is inverted, (21/34) f behind (right of) lens, inverted, and
5 13 34
1/17 original size.
2-23. The arrangement of the object and lenses is shown below.
20 cm 30 cm 20 cm
1 2 3
(a) f1 = + 10 cm, f2 = + 15 cm, f3 = + 20cm
1 1 1
1st lens: + = s ′ = 20 m1 = − 20/20 = − 1
20 s ′ 10
1 1 1 ′
2nd lens: + = s = − 30 m2 = − ( − 30)/10 = + 3
10 s ′ 15
1 1 1
3rd lens: + = s ′ = 100/3 m3 = − 100/3 (50) = − 2/3
50 s ′ 20
mT = m1 m2 m3 = + 2
(b) f1 = + 10 cm, f2 = − 15 cm, f3 = + 20cm
1 1 1
1st lens: + = s ′ = 20 m1 = − 20/20 = − 1
20 s ′ 10
1 1 1
2nd lens: + = s′ = − 6 m2 = − ( − 6)/10 = + 0.6
10 s ′ − 15
1 1 1 10
3rd lens: + = s ′ = 520/6 m3 = − 520/(6 × 26) = −
26 s ′ 20 3
mT = m1 m2 m 3 = + 2
(c) f1 = − 10 cm, f2 = + 15 cm, f3 = − 20cm
1 1 1 1
1st lens: + = s ′ = 20/3 m1 = − ( − 20)/3 (20) =
20 s ′ − 10 3
3 1 1 ′ (330) (3) 9
2nd lens: + = s = 330/13 m2 = − =−
110 s ′ 15 (13) (110) 13
− 13 1 1 ′ (140) (13) 26
3rd lens: + ′= s = 140/19 m3 = − =
70 s − 20 (19) ( − 70) 19
mT = m1 m2 m3 = − 6/19

1 n − n1 1 1
2-24. Using the lensmaker’s formula, = 2 − gives
f n1 R1 R2

1 1.50 − 1 1 1
in air: = −
30 1 R1 R2
1 1.50 − nL 1 1
and in the liquid: − = − .
188 nL R1 R2
− 188 0.5nL
Dividing the two equations gives, = or nL = 1.63.
30 1.5 − nL
7
2-25. Use the lensmaker’s formula to find the focal length of the lens,

1 n2 − n1 1 1 1.5 − 1 1
= − = 0+ ⇒ f = 120 cm
f n1 R1 R2 1 60
f x′
The Newtonian equations are, m = − = − . For m = − 4, .
x f
f 120
−4=− =− or x = 30cm
x x
x′ x′
−4=− =− or x ′ = 480cm
f 120
Thus, s = x + f = 30 + 120 = 150cm and s ′ = x ′ + f = 480 + 120 = 600cm.
1 1 1 1 1 1
Check: + = + = =
s s ′ 150 600 120 f
1 1 1
2-26. (a) f1 = 10 cm ⇒ P1 = = + 10 D, f2 = 20 cm ⇒ P2 = = + 5 D, f3 = − 40 cm ⇒ P3 = = − 2.5 D
0.01 0.2 − 0.4
Then, P = P1 + P2 + P3 = 10 + 5 − 2.5 = + 12.5 D
1 1 1 ′ 1 1
(b) + ′ = , V + V = P , where V = = = + 8.33 D,
s s f s 0.12
′ 1 1
V = 4.167 D or s ′ = ′ = = 0.24 m = 24cm
V 4.167
2-27. See Figure 2 38 in the text. The applicable relations are:

1 1 1 1 1 1
Lens equations: + ′ = and + ′= ,
s1 s1 f s2 s2 f
Geometrical: L = s1 + s1′ = s2 + s2′ , D = s2 − s1 = s1′ − s2′
Thus,
s1 s1′ s s′ s s′ s s′
f= ′ = 1 1 = 2 2′ = 2 2 (1)
s1 + s1 L s2 + s2 L
Because the lens equation can be satisfied the second time by simply interchanging object and image dis
tances,
s2 = s1′ and s2′ = s1 (2)
Adding and subtracting the equations L = s2 + s2 and D = − s1 + s2, we get,
L − D = 2s and L + D = 2s2. Their product is by Eq. (1), L2 − D2 = 4 s1 s2, or by Eq. (2), L2 − D 2 = 4 f L.
L2 − D 2
Thus, f = .
4L
1 1 1 1 1 1
2-28. Lens equations: + = and + = . Then calculate,
s1 s1′ f s2 s2′ f
1 1 s s s1 s2 s − f s1 − f s2 − s1
− = − 1′ + 2′ = − + = 2 − =
M1 M2 s1 s2 s1 f /(s1 − f ) s2 f /(s2 − f ) f f f
Thus,
s2 − s1
f=
1/M1 − 1/M2
2-29. Consider an arbitrary path from point A to point B by reflection from a mirror surface,
√ p
The path distance D from A to B is D = a2 + x2 + b2 + (d − x)2
A
B dD x − (d − x)
θ i θr =√ +p =0
a a d x a 2 + x2 b2 + (d − x)2
x
sin θi − sin θr = 0 ⇒ θi = θr
d
8
2-30. The two set ups are illustrated below,
1 n n−1 ′ nR
Refraction at curved side: + = ; s −
∞ s′ R n−1
nR
Reflection at plane side: s ′ = − s = −
n−1
n 1 1−n
Refraction at curved side: + = or s ′ = R/2 (n − 1)
− n R/(n − 1) s ′ −R
R
Thus, f1 =
2 (n − 1)
1 1 2 R
Reflection at curved face: + = − ; s′ = −
∞ s′ R 2
n 1 ′ R
Refraction at plane face: + = 0; s =
− R/2 s ′ sn
n 1 R
Refraction at plane face: + ′ = 0; s ′ =
− R/2 s sn
Thus, f2 = R/2 n
f1 R/2 (n − 1) n
Therefore the ratio of the focal lengths is = = .
f2 R/2 n n−1
fs
2-31. The distance between the object and the image is D = s + s ′ = s + . This is minimized when,
s−f
dD (s − f ) f − f s
=1+ = 0 ⇒ s (s − 2 f ) = 0 ⇒ s = 0, 2 f . The minimum distance D occurs when s = 2f
ds (s − f )2
f (2f )
and has the value D = 2f + = 4f . That is, in this configuration s = s ′ = 2f .
2f − f

(a) Let the angle with the normal to the interface in each region of index of refraction ni be θi. Then
applying Snell’s law sequentially at each interface leads to,
n0 sinθ0 = n1 sinθ1 = n2 sinθ2... = ni sinθi... = nf sinθ f
That is,
n0 sinθ0 = n f sinθ f
(b) In each medium the lateral displacement is ti tanθi. The total lateral displacement y due to N media
can be written as,
X n
y= ti tanθi
i=1
where sinθi = (n0/ni) sinθ0.
2-33. At each surface use the relation,

n1 n2 n2 − n1
+ ′=
s s R
For light incident first on the plane side:
1.5 1 1 − 1.5
1st surface (plane): no change, 2nd surface (curved): + = or s ′ = 8 cm.
∞ s′ −4
For light incident first on the curved side:
1 1.5 1.5 − 1
1st surface (curved): + ′ = or s ′ = 12cm.
∞ s 4
1.5 1 1 − 1.5
2nd surface (plane): object distance = 4 − 12 = − 8 cm (virtual), + = or s ′ = 5.33cm
− 8 s′ ∞
9
2-34. The focal length is the image position for incident parallel light rays (object at ∞). In all cases the fol
lowing relation is to be used
n1 n2 n2 − n1 n n n − n1 n n − n1
+ ′= ⇒ 1+ 2= 2 ⇒ 2= 2
s s R ∞ f R f R
For the situation in which the center of curvature in medium with n = 4/3:
4/3 4/3 − 1
For light incident from the medium of index 1: = or f = + 40cm
f 10
1 1 − 4/3
For light incident from the medium of index 4/3: = or f = + 30cm
f − 10
For the situation in which the center of curvature is in the medium with n = 1,
4/3 1 1 − 4/3
For light incident from the medium of index 4/3: + = or f = − 30cm
∞ f 10
1 4/3 4/3 − 1
For light incident from the medium of index 1: + = or f = − 40cm
∞ f − 10
s 1 f 6 in
2-35. |m| = = = = , since s ′ = f . So s = 50, 000 × 6 in = 25, 000ft
s ′ 50, 000 s s
2-36. Using, the lensmaker’s equation the focal power of the cylindrical lens is,
1 n2 − n1 1.60 − 1.0
= = ⇒ f = 8.33 cm
f R 5 cm
The image distance is then found as,
1 1 1 1 1
= − = − ⇒ s ′ = 18.73 cm
s ′ f s 8.33 cm 15 cm
Then, Eq. (2 37) gives,
s + s′ 15 + 18.73
AB = CL = 7 cm = 15.75 cm
s 15
The line image is real, 18.75 cm past the lens and 15.75 cm long.
1 n2 − n1 1.52 − 1.0
= = ⇒ f = 28.85 cm
f R 15 cm
1 1 1 1 1
The image distance is then found as, = − = − ⇒ s ′ = − 65.2 cm
s ′ f s 28.85 cm 20 cm
s + s′ 20 − 65.2
AB = CL = 2.5 cm = − 5.65 cm
s 20
The line image is virtual, 65.2 cm from the lens on the object side of the lens and 5.65 cm long.
10
1 n2 − n1 1.5 − 1.0
= = ⇒ f = − 20 cm
f R − 10 cm
1 1 1 1 1
= − = − ⇒ s ′ = − 11.11 cm
s ′ f s − 20 cm 25cm
s + s′ 25 − 11.11
AB = CL = 5 cm = 2.78 cm
s 25
1 n2 − n1 1.5 − 1.0
= = ⇒ f = − 40 cm
f R − 20 cm
1 1 1 1 1
= − = − ⇒ s ′ = − 13.33 cm
s ′ f s − 40 cm 20cm
s + s′ 20 − 13.33
AB = CL = 2 cm = 0.667 cm
s 20
1 n2 − n1 1.60 − 1.0
= = ⇒ f = 8.33 cm
f R 5 cm
1 1 1 1 1
= − = − ⇒ s ′ = − 21.45 cm
s ′ f s 8.33 cm 6 cm
s + s′ 6 − 21.45
AB = CL = 7 cm = − 18.0 cm
s 6
11
Chapter 3 Optical Instrumentation
3-1. The entrance pupil is the aperture stop so no elements precede the aperture stop. The exit pupil is the
image of the aperture stop formed by the lens. The position and size of the the exit pupil are found from
the thin lens equation with the object being the aperture stop.
Exit Pupil and Size:
1 1 1 1 1 1
+ = ⇒ + = ⇒ s ′ = − 3.33 cm
s s′ f 2 cm s ′ 5 cm
s′ 3.33 5 5
m=− =− = ⇒ size = × 2 cm = 3.33cm
s 2 3 3
Image Position and Size:
1 1 1
+ = ⇒ s ′ = 10cm
10 cm s ′ 5 cm
m = − s ′/s = − 10/10 = − 1 ⇒ size = 1 × 2 cm = 2 cm, inverted
The system is drawn to scale below.

ExP AS Lens
Image
Object
3-2. Exit Pupil: AS (no elements follow), Entrance Pupil: Image of AS formed by lens
1 1 1
Position of EnP : + = ⇒ s ′ = − 4.29 cm = − 30/7 cm (right of lens)
2.5 cm s ′ 6 cm
− 30 2 12 12
Size of EnP : m = − s ′/s = − × = size = × 2 cm = 3.43cm
7 5 7 7
1 1 1 3
Image: + = ⇒ s ′ = 10.5cm, m = − s ′/s = − 10.5/14 = − 0.75; size = × 4 cm = 3 cm, inverted
14 cm s ′ 6 cm 4
En P
AS
Lens
Image
Ob ject
12
3-3. Entrance Pupil: AS (no preceding elements), Exit Pupil: Image of AS through lens
1 1 1 − 12
Exit Pupil: + = ⇒ s ′ = − 12cm. m = − s ′/s = − = + 3, size = 3 × 2 cm = 6cm
4 cm s ′ 6 cm 4
1 1 1
Image: + = ⇒ s ′ = 10.5 cm , m = − s ′/s = − 10.5/14 = − 0.75, size = 3/4 × 2 = 1.5cm
14 cm s ′ 6 cm
ExP
AS
Lens
Image
Ob ject
3-4. (a) The vertical and horizontal scales in the figure differ by a factor of 4.
Object plane Image plane
Lens 1 Lens 2
Aperture
(b) Image Plane:

f 1 s1 (40/3) (40) f s (20/3) (10)
s1′ = = = 20 cm (right of Lens 1) s2′ = 2 2 = = 20 cm (right of Lens 2)
s1 − f1 40 − 40/3 s2 − f 2 10 − 20/3
(c) Candidates for AS: Lens1, Aperture, or Lens 2.

2
Lens subtends at an angle of θL1 = = 0.05 rad.
40
20 (40/3)
Aperture image in Lens 1: s ′ = = 40 cm, or at object plane, then θAperture = 0◦
20 − 40/3
30 (40/3)
Lens 2 image through Lens 1: s = = 24 cm left of Lens 1 or 16 cm right of object.
′
30 − 40/3
1.6
m = − s /s = − 24/30 = − 0.8 so size = 0.8 × 2 = 1.6cm, θL2 = = 0.1 > 0.05
′
16
Thus Lens 1 behaves as the AS. It is also the EnP , being the first in line.
(d) Exit Pupil Ex P: Image of AS (i.e., Lens 1) in Lens 2:
30 (20/3) 8.57
′
sf 2 s 2 4
= 8.57 cm, right of Lens 2, m = − = − = 0.2857 = 2/7 so DExp = × 2 = cm
′
s= =
s − f 2 30 − 20/3 s 30 7 7
(e) Field Stop: Either Aperture or Lens 2, whichever subtends the smaller angle at center of EnP =Lens
1:
0.5 2
θA perture = = 0.025; θLens 2 = = 0.067
20 30
Thus the Aperture Stop behaves as the field stop (FS). Continued on next page
13
3-4. (e) continued .
Exit Window: Image of FS in following elements lens Lens 2.
(10) (20/3)
= 20 cm, right of Lens 2, in image plane,
′
s=
10 − 20/3
m = − s /s = 20/10 = − 2 so DExω = 2 × 0.5 = 1.0cm
′
(f) Angular field of view: Drawn not to scale,
En W En P ExP ExW
1 1
1 cm α α= = 0.0125 rad α′ 1 cm α′ = = 0.0875 rad
40 10 − 8.43
α = 1.43◦
α ′ = 5.01◦
40 cm 11.43 cm

(a) First determine whether A, L1, or L2 (its image though L1) subtends the smallest angle at he center of
EnP :
Rim of A at EnP : 90◦. Rim of L1 at EnP : tan−1(3/3) = 45◦.
Rim of the image of L2 in L1:
This image is labeled L2′ in Figure 3 5 in the text and is shown there to be virtual, erect, located 15 cm
from EnP with half size 9 cm. Thus L2′ subtends an angle of tan−1(9/15) = 31◦ at EnP . So L2′ subtends
the smallest angle at the entrance window making L2 the field stop, of half size 3 cm.
(b) The entrance window is the image of the field stop in all optics to its left. Since the field stop is L2, L2′
is the entrance window. As shown in Figure 3 5 in the text it is located 12 cm to the right of L1 with half
size 9 cm.
The exit window is the image of the field stop in all optics to its right. Since, in this case there are no
optics to the right of the field stop (L2), the exit window is L2 of half size 3cm.
(c) The two marginal rays can be sketched onto Figure 3 5 in the text as follows. The marginal rays start
from point P and skim the edges the entrance pupil A. Upon encountering L1 the rays are head towards
P ′. Before they reach P ′ they are redirected towards the tip of the final image at P ′′. The backward exten
sion of the marginal rays heading toward P ′′ skim the edges of the exit pupil.
3-6. The deviation δ is given as a function of the input angle θ1 by Eqs. (3 7), (3 8), (3 10), and (3 11) as,
δ = θ1 + θ2 −θ1′ − θ2′ = θ
1 + θ2 − A
−1 sin(θ1)

δ = θ1 + sin −1
n sin A − sin −A
n
For n = 1.52 and A = 60◦, I obtain the plot below.
45
44
43
42
δ
41
40
39
38
35 40 45 50 55 60 65 70
θ1
14
3-7. The index of refraction is given by Eq. (3 15)

A + δm
sin 2 sin (30 − δm/2)
n= =
sin (A/2) 1/2
1
For nred = 1.525 sin (30 + δm/2) = (1.525) or δm,red = 39.37◦
2
1
For nblue = 1.535 sin (30 + δm/2) = (1.535) or δm,blue = 40.26◦
2
δm,blue − δm,red = 0.89◦ = 53 ′
3-8. The necessary data is
n (crown) n(flint)
F (481.6 nm) 1.5286 1.7328
C (656.3 nm) 1.5205 1.7076
(a) The Cauchy relations for λF and λC are: nF = A + B/λ2F , nC = A + B/λ2C . Solving these simultaneously,
B nF − nC
A = nF − and B =
λ2F (1/λ2F − 1/λ2C )
4240 1.5286 − 1.5205

For crown glass: A = 1.5286 − = 1.511 B= = 4240nm2
466.12 1
−
1
(486.1)2 (656.3)2
1.319 × 104 1.7328 − 1.7076

For flint glass: A = 1.7328 − = 1.677; B = = 1.319 × 104 nm2
486.12 1
2 −
1
2
(486.1) (656.3)
Then for a third wavelength λD = 589.2 nm,
B 4240
Crown: nD = A + 2 = 1.511 + (589.2)2 = 1.523 , (table: 1.5230)
λD
1.319 × 104
Flint: nD = 1.677 = = 1.715 , (table: 1.7205)
(589.2)2
dn 2B
(b) The dispersion is D = = − 3 so,
dλ λ
− 2(4240nm2) − 2 (13190nm2)
DC = = − 4.146 × 10−5 nm −1 DF = = 1.290 × 104 nm−1
(589.2 nm) 3 (589.2 nm)3
(c) The resolving powers are,
RC = b DC = (7.5 × 107 nm) (4.146 × 10−5nm −1) = 3110

RF = b DF = (7.5 × 107 nm) (1.290 × 10 −4 nm−1) = 9675
The minimum resolvable wavelengths are,
crown: (∆λ)m in = λ/RC = 5892 Å/3110 = 1.9Å

flint: (∆λ)min = λ/RF = 5892 Å/9675 = 0.61 Å
15
sin (30 + δ/2)
3-9. (a) 1.6381 = or δ = 50◦
sin 30
nF − nC 1.63461 − 1.64511 1
(b) ∆ = = =−
nD − 1 1.6381 − 1 55.5
n1 − n2
(c) D = − 2B/λ3 where B = 1 1 , as in problem 6 8a. For the long wavelength region, take λ1 = 587.6 nm
λ21
− λ2
2
1.63810 − 1.63461 B 6073.7
and λ2 = 656.3 nm. Then B = = 6073.7nm2 and A = n1 − 2 = 1.63810 − =1.6205.
1/587.62 − 1/656.32 λ 587.62
− 2 (6073.7)
Finally, D(at λ = 656.3 nm) = = 4.297 × 10−5 nm−1
656.33
λ
(d) The resolving power is R = AV = b D . Now ∆λ = 656.2852 − 656.2716 = 0.0136 nm, so that
∆λ
λAV 656.28nm
b= = =1.12 × 10 nm = 1.12 m!
D ∆λ (4.297 × 10−5 nm−1) (0.0136 nm)
3-10. The indices of refraction of the lines are,
sin [(60 + 38◦20 )/2]

′
nC − = 1.51323. Similarly, nD = 1.51570 and nF = 1.52308

sin (60/2)
nF − nC
Then, ∆ = = 0.01909
nD
3-11. Let the flint prism be prism (1) and the crown be prism (2). Then, dispersion of the flint prism is δD1 =
(1.635 − 1) (5◦) = 3.175 . This dispersion must be matched by the crown prism, δD2 = 3.175 = (1.53 − 1) A2,
◦ ◦
or A2 = 5.99◦. For the C and F lines with this double prism,
δF 2 = (1.536 − 1) (5.99) = 3.211◦ δF 1 = (1.648 − 1) (5) = 3.240◦ Net: δF 2 − δF 1 = 0.029◦

δC2 = (1.527 − 1) (5.99) = 3.157◦ δC1 = (1.6301 − 1) (5) = 3.150◦ Net: δC2 − δC1 = − 0.007◦
Dispersion = δF 12 − δC12 = 0.029 − ( − 0.007) = 0.036◦ = 2.16 ′
3-12. Let the crown prism be prism 1 and the flint prism be prism 2. Then,
δC1 = (nC1 − 1) A1 = (1.5205 − 1) 15 = 7.8075◦

δF 1 = (nF 1 − 1) A1 = (1.5286 − 1) 15 = 7.929◦
The relative dispersion is δF 1 − λC1 = 0.1215◦. This must be reversed by the flint prism:
δF 2 − δC2 = 0.1215◦ = (nF 2 − nC2) A2 = (1.7328 − 1.7076) A2 ⇒ A2 = 4.82◦
The deviation of line D is then, δD1 − δD2 = (nD1 − 1) A2 = (.523) (15) − (0.7205) (4.821) = 4.37◦
16
Φtot 25 W
3-13. (a) The radiant exitance is given by Me = = = 104 W/m2.
A (0.05 m)2
Since the source is perfectly diffuse (i.e. is Lambertian) the radiant intensity can be written in the form,
Ie = Ie(0) cos θ. Here θ = 0 is the direction normal to the surface of the square source. Take the θ = 0 direc
tion to be along the positive z axis. The total power emitted in the forward direction is related to the
intensity via the relation,
Z Z 2π Z π/2
1
Φtot = Iedω = dϕ Ie(0) cos(θ)sin(θ)dθ = 2 π I(0) = π Ie(0)
0 0 2
Thus,
Φtot 25 W
Ie(0) = = = 7.96 W/sr
π sr π sr
This is the radiant intensity in the direction of the surface normal to the source. Since the source is Lam
bertian the radiance Le is independent of the viewing angle θ and given by,
I(0) cos θ I(0) 7.96 W/sr W
Le = = = = 3180 2
A cos θ A (0.05 m)2 m · sr
(b) Assume that the camera aperture is parallel to the source surface and centered along the direction of
the surface normal. Now, f /D = 8 = (4 cm)/D. Thus, D = 0.5 cm. The flux Φcam entering the camera is
well approximated by multiplying the intensity in the forward direction by the solid angle intercepted by
the camera lens. Noting that the source to lens distance is r = 1 m, this is, approximately,
!
π (0.25 cm) 2
Φcam = I(0) ∆ωlens = I(0)(Alens )/r 2 = (7.96 W/sr) = 1.56 × 10 −4 W
(100 cm)2
Φcam Φ
(c) The irradiance at the film is then, Ee = = cam
Aimage |m|2 A
Here, we have noted that each image dimension is decreased from the source dimension by the magnifica
tion,
s′ f 4
|m| = = = = 0.0417
s s − f 100 − 4
So,
1.56 × 10−4 W
Ee = = 35.9 W/m2
0.0417)2(0.05 m)2
3-14. Plots as a function of the different parameters are shown below.

18 16 16
16 14 14
Depth of F e d (m)
Depth of F e d (m)
Depth of F e d (m)
14 12 12
12 10 10
10
8 8
8
6 6 6
4 4 4
2 2 2
0 0 0
3 4 5 6 7 8 0 5 10 15 20 0 2 4 6 8 10 12
f (cm) A s0 (cm)
A = 8, s0 = 2 m, d = 0.05mm f = 5 cm, s0 = 2 m, d = 0.05 mm A = 2, f = 5 cm, d = 0.05 mm
3-15. From Eqs. (3 30) and (3 31),

A s0 d (s0 − f ) (4) (6) (10−6) (6 − 0.05)
s0 − s1 = = m = 0.057 m = 5.7cm
f 2 + A s0 d (.05)2 + (4) (6) (10−6)
A s0 d (s0 − f )
s2 − s 0 = ≈ s0 − s1 = 5.7cm
f 2 − A s0 d
Thus, points at 5.7 cm nearer or farther than the middle row create blur circles larger than the silver grain
dimension. Under these demanding conditions, the picture could not be taken successfully.
17
3-16. Using Eq. (3 35),
1 1 1 L 1 1 15
= + − = + − ⇒ f = 53.3cm
f f1 f2 f1 f2 20 − 8 (20) ( − 8)
If the film plane is in the image position for an object at infinity, its position can be found as follows. The
image formed by the first lens of the object at infinity falls at the focal point of the first lens which is 5 cm
past the second lens. This intermediate image serves as the object for the second lens. That is s2 = − 5 cm.
The final image position relative to the second lens can be found using the thin lens formula,
1 1 1 1 1 1
+ ′= ⇒ + = ⇒ s2′ = 13.33 cm
s2 s2 f 2 − 5 s2 − 8
That is, the film should be placed 13.33 cm past the negative lens of the combination. The image size h ′ of
a distant object subtending an angle of 2◦ at the camera is found using the effective focal length of the lens.
|h ′| h′
tan 2◦ = = ⇒ 1.86 cm
f 53.33 cm
Alternately one may find the image height as,
s′ s′ 20 cm 13.33

h
h ′ = m1 m 2 h = − 1 − 2 = h = (53.33 cm) = (53.33 cm) tan 2◦ = 1.86 cm
s1 s2 s1 −5 s1
3-17. Using the parameters f = 5 cm = 0.1640ft, s0 = 6ft, A = f /4 = 4, and d = 0.05 mm = 1.640 × 10−4 ft in Eqs,
(3 30) and (3 31) gives,
s0 f (f + A d)
s1 = = 5.26ft
f 2 − A d s0
s f (f − A d)
s2 = 0 2 = 7.00ft
f − A d S0
3-18. See Figure 3 38 in the text. The diameter of the image of the sun is about
π rad

DI = fL θ = (50 cm) (0.5◦) = 0.44 cm. Since all of the power intercepted by the lens, falls on the
180◦
image, the irradiance of the image is
2
2
Aimage DL 5
= Isun 2 = 1000 W/cm 2
= 1.29 × 105 W/cm2

Iim age = Isun
Alens DI 0.44
3-19. (a) Using f = 15 cm, s = 100 ft = 3048 cm, and h = 6 ft, one finds,
− f − 15
′
s
m=− = =
s s 3048
15
′
h = |m| h = 6 ft = 0.0295 ft = 0.9 cm
3048
(b) Distant objects form an image in the first lens at f1 = 12 cm. Thus, the object distance for lens 2 is
s2 = (8 − 12) cm = − 4 cm . Then the image distance from lens 2 is s2′ = 15 cm so the focal length of the
second lens must be,
1 1 1 1 1
= + = + ⇒ f2 = − 5.45 cm
f2 s2 s2′ − 4 cm 15 cm
Then for the person at 100 feet,
′
! ′
!
− 12 − 15

s s
m = m1 m2 = − 1 − 2 = − = − 0.0148
s1 s2 3048 −4
The ratio of the magnifications is

|m| 15/3048 1
= =
|m(telephoto)| 180/4 (3048) 3
18
f 50mm
3-20. (a) D = = = 27.8mm
1.8 1.8
√
(b) Factor of 3 : f /1.8 → f /3.1 → f /5.4 → f /9.4
(c) D = 50/3.1 = 16.0 mm; D = 50/5.4 = 9.26 mm; D = 50/9.4 = 5.35mm
1
(d) f /1.8 at is equivalent in exposure to f /3.1 at 3/100 s, f /5.4 at 9/100 s, f /9.4 at 27/100 s
100
25 1 1 1 L
3-21. M = where = + − and L is determined by the condition for elimination of chromatic
Feq feq f1 f2 f1 f2
1
aberration: L = (f1 + f2). Then,
2
25

1 1 1 f + f2 1 1 1 1 1
= + − 1 = + and M = = 12.5 +
feq f1 f2 2f1 f2 2 f1 f2 feq f1 f2
3-22. The focal lengths of the two lenses are, f1 = f2 = 3 cm and they are separated by L = 2.8cm.
(a) The equivalent focal length is,
1 1 1 L 1 1 2.8
= + − = + − feq = 2.8125cm
feq f1 f2 f1 f2 3 3 9
(b) The magnification is,
25 25
M= +1= + 1 = 9.9 ≈ 10 ×
feq 2.8125
3-23. (a) The system is sketched below.

Obj Ocular s′
Fo b j M = mob j × Mo c = × 10 where s ′ = f + 16 = 16.5cm
s
(16.5) (0.5) 8.25
′
s f
s= ′ = =
16 cm s −f 16.5 − 0.5 16
16.5 cm s
′
(16.5) (16)
|m| = = = 32, M = (32) (16) = 320 ×
s 8.25
8.25
(b) s = = 0.516 cm
16
′
sob j 25

s f (1.20) (1)
+ 1 where, sob j = 0 ob j =
′
3-24. (a) M = mob j × Mo c = = 6cm. So,
sob j fo c s0 − fob j 1.20 − 1
25

6
M= + 1 = 46.7 ×
1.20 3
′
s f ( − 25) (3)
(b) L = sob j + so c = 6 + 2.68 = 8.68 cm where so c = ′ o c o c =
′
= 2.68cm
so c − f o c − 25 − 3
25 25 25 − 4 − fo

L
3-25. M = − and L = d − fo − fe ⇒ − 20 = − ⇒ fo = 5cm
fe fo 4 fo
0.2013 0.001
′ ′
1 1 1 s h
⇒ s = 0.2013 m, m = = = 0.006711, h = m h ⇒ h = = = 0.149 m = 14.9 cm
′ ′
3-26. + =
30 s′ 0.2 s 30 m 0.006711
19
fob j 14 D 35
3-27. (a) M = 7 × (b) M = or fo c = = 2 cm (c) M = ob j or Dex = = 5mm
fo c 7 Dex 7
sf L foc (14 + 2) (2) 16
(d) The exit pupil position is the image of L1 in L2, or s = = 2.3 cm
′
= = =
s − f L − foc (14 + 2) − (2) 7
(e) Let y be the linear dimension of the field of view at a distance of x = 1000 yd. Then the angular field of
view can be written as,
Df 1.8 y
θ= = = 0.1125 rad = ⇒ y = x θ = (1000 yd)(0.1125 rad) = 337 ft.
L 14 + 2 x
The field of view is 337 ft at 1000 yd. The geometry is sketched (not to scale) below.
Obj
y θ θ Df
x L
3-28. (a) Consider the right triangles in the figure below that share the image y as a common size. Form these
two triangles one can write,
tan α ′ fob j
y = fob j tan α = so c tan α ′ ⇒ =
tan α so c
α ′ tan α ′ fob j
The angular magnification is then, M = ≈ =
α tan α so c
Since mo c = − s /so c , the required result follows,
′′
fob j mo c
M =−
s ′′
O b jective O cular
fo b j so c
α α′
α α′
O b ject
s ′′
Final im age
(b) When viewed at ∞, M = fo/fe = 30/4 = 7.5 × . When viewed at 25 cm, s ′′ = − 25 cm and so,
s ′′ s ′′ s ′′ − fo c ( − 25) − 4 29
mo c = − = − ′′ ′′
=− =− = = 7.25 ×
so c s fo c /(s − fo c ) fo c 4 4
Them by the formula from part (a),
(30) (7.25)
M =− = 8.70 ×
( − 25)
3-29. The image size on the moon is determined by the angle α ′ shown in the figure associated with the solution
to problem 3 28 above. In that problem solution it is shown that M = α ′/α. Here α = 0.5◦ = π/360. The
diameter of the image of the moon is given by DM o on = s ′′ α ′ = (25 cm) α ′. To find α ′ use the relation
obtained in the solution to problem 3 28), M = fob j /so c . Here, from a rearrangement of the thin lens for
mula,
s ′′ f ( − 25) (5) 25
so c = ′′ o c = cm = cm
s − fo c − 25 − 5 6
so that,
20 24
M= = ×
(25/6) 5
Then,
24 π π π π
α′ = M α = = and DM o on = s ′′ α ′ = 25 cm = cm = 1.05 cm
5 360 75 75 3
20
3-30. (a) L = fob j + fo c = 12 cm − 4 cm = 8 cm. M = fob j/fo c = (12/4) = 3 × .
(b) For the image at s ′′ = 30 cm:
s ′′ f − 30 ( − 4) 60
so c = ′′ o c = cm = − cm
s − fo c − 30 + 4 13
f 12
M = ob j = = 2.6 ×
so c 60/13
L = fob j + so c = 12 cm − (60/13) cm = 7.38 cm
3-31. See Figure 3 39 in the text. In the normal position, so c = fo c = 5 cm. But when s ′′ = 25 cm,
s ′′ fo c 25 · 5
so c = = cm = 6.25 cm
s ′′ − fo c 25 − 5
Thus, the ocular must be moved further from the objective by an amount 6.25 cm − 5 cm = 1.25 cm in order
to produce a real image on the screen 25 cm away.
3-32. Using Eq. (3 35) and the given information,

1 1 1 L 1 1 2
= + − = + − ⇒ feq = 2 cm
feq f1 f2 f1 f2 2 cm 2 cm 4 cm
25 25 f 30
(a) M∞ = = = 12.5 × (b) Mtel = ob j = = 15 ×
feq 2 fo c 2
(c) The exit pupil is the image of the AS or objective lens, formed by the ocular. Since the ocular consists
of 2 lenses, one should argue as follows: To produce parallel rays leaving the ocular (image at ∞), the field
lens must be at the focal point of the objective, 30 cm from the objective. Then the image there is 2 cm
from the eye lens (1/2 + 1/s ′ = 1/2, s ′ → ∞). Given this separation, the image of the objective in the field
lens is
30(2) 15
s1′ = cm = cm
30 − 2 7
that is 1/7 cm beyond the eye lens. The image formed by the eye lens is then at,
( − 1/7) (2)
s2′ = cm = 0.133 cm
− 1/7 − 2
Thus the exit pupil falls 0.133 cm from the eye lens.
4.5
Using M = Dob j /DExP we also have DExP = cm = 3mm.
15
(d) θ = DFL /L = 2/30 = (1/15) rad = 3.8◦.
3-33. “Unfolding” the right angle in the optical axis of Figure 3 34a in the text, the equivalent optics shown in
Figure 3 40 associated with the problem in the text is seen. From the diagram in Figure 3 40,
αm h/fe fo
|M | = = =
αo h/fo fe
3-34. Refer to figure 3 41 in the text. For M2:

1 1 1 s s′ ( − 2)(10)
+ ′= ⇒f= = ft = − 2.5 ft
s s f s+s ′ − 2 + 10
Without the intermediate convex mirror, the optical system is the same as a Newtonian telescope, whose
magnification is given (see problem 3 33) by
fob j 12
M =− =− = − 36 ×
fo c − 4/12
The magnification is increased by the convex mirror by the factor, m = − s ′/s = − ( − 10)/( − 2) = + 5
The overall magnification is then, Moverall = m M = (5)( − 36) = − 180 ×
21
Chapter 4 Wave Equations
2
4-1. The waveform is sketched below. As a function of time y = A e−b(x+10 t)
y
4-2. (a) Replace x by x + v t:

4 m3
y(x, t) =
(x + (2.5 m/s))2 + 2 m3
(b)
2 t=0 s
t=2 s
t=5 s
1.5
y(m)
0.5
0
-20 -15 -10 -5 0 5 10 15 20
x(m)
4-3. (a), (b)

Wave form 1 is a traveling wave since it is a function of u = z − v t with v = − 1 m/s. This means that it
travels in the negative z direction with a speed of 1 m/s.
Wave form 2 is a traveling wave since it is a function of u = x − v t, with v = 1 m/s. This is a wave traveling
in the positive x direction with a speed of 1 m/s.
Wave form 3 is not a traveling wave since it is not a function of simply u = x − v t.
With more labor one can show that waveforms 1 and 2 do and waveform 3 does not satisfy the differential
wave equation (Eq. (4 2)).
4-4. The waveform can be rewritten as,

(100 m2 ) x−(10 m /s)t)2/m 2
y= e
x − (10 m/s) t
Thus y = y(x − (10 m/s) t) and so the waveform is a traveling wave with vx = 10 m/s.
4-5. From the given information, k = 2 π/λ = (2 π/5) m , |v| = f λ = λ/T = (5/3) m/s, vz = − 5/3 m/s
−1
z t
(a) y = A sin 2 π(z/λ ± t/T ) = 2 sin 2 π +
5m 3s

2π 5
(b) y = A sin k(z ± v t) = 2 sin z+ m/s t
5m 3
(c) ỹ = A e2πi(z/λ±t/T ) = 2 e2πi[z/(5m )+t/(3s)], y = Im(ỹ )
4-6. (a) y = A sin 2 π(x/λ ± t/T ) = (5 m)sin[2 π x/(50 m)]

2p h π i
(b) y = A sin k (x ± vx t) = (5 m) sin (x + (2 m/s) (4 s)] = (5 m) sin (x + 8 m)
50 m 25 m
22
628.3 x 6283

x t
4-7. Compare the general form, y = sin 2 π − to the given form, y = (10 cm) sin − t
λ T cm s
(a) 2 π/λ = 628.3/cm ⇒ λ = 0.01 cm
(b) ν = 1/T ; 2 π/T = 6283/s ⇒ T = 0.001 s, ν = 1000 Hz
(c) k = 2 π/λ = 628.3/cm (d) ω = 2 πν = 2 π(1000)/s = 6283/s (e) T = 0.001 s
(f) v = ν λ = (1000) (0.01) cm/s = 10 cm/s (g) A = 10 cm
dϕ dy dy
4-8. (a) ϕ = y − B t = constant, =0= − B ⇒ vy = = B. (B has units of m/s )
dt dt dt
dϕ dx dx
(b) ϕ = B x + C t + D = constant, =0=B + C ⇒ vx = = − C/B.
dt dt dt
(B x/C t is unitless so C/B in m/s )
√ √

dϕ dz
(c) ϕ2 = B z 2 + B C 2 t2 − 2 B C z t = B (z − C t)2, ϕ = B (z − C t)=constant, =0= B −C .
dt dt
dz
Therefore, vz = = C. Apparently C t and the same dimensions as z so C has dimensions of m/s .
dt
4-9. Taking the general form at t = 0 as y = A sin [2 π (x/λ) + ϕ], the given conditions are,
y(x = 0) = 13 = A sin ϕ (1)
y(x = 3 λ/4) = − 7.5 = A sin (1.5 π + ϕ) = − A cosϕ (2)
Dividing (1) by (2) gives,
− 13/7.5 = − tan ϕ ⇒ ϕ = tan−1(13/7.5) = 1.047 ≈ π/3
Then, from (1)
A = 13/sin ϕ = 13/sin(π/3) = 15.0
So that,
y = 15 sin(2 π x/λ + π/3)
4-10. (a) Take the waveform at t = 0 to be y = A sin(2 π x/λ + ϕ0). We are given that y = A at x = x0. That is,
y(x0) = A sin(2 π x0/λ + ϕ0) = A
sin(2 π x0/λ + ϕ0) = 1
2 π x0/λ + ϕ0 = π/2
π 2π
ϕ0 = − x0
2 λ
(b) For λ = 10 cm,
x0 = 0, ϕ0 = π/2
π
x0 = 5/6 cm, ϕ0 =
3
x0 = 5/2 cm, ϕ0 = 0
π
x0 = 5 cm, ϕ0 = −
2
3π
x0 = − 1/2 cm, ϕ0 =
5
(c) Generally y = A cos(2 π x/λ + ϕc) the same argument given in (a) requires that
cos(2 π x0/λ + ϕc) = 1
2 π x0/λ + ϕc = 0
ϕc = − 2 π x0/λ = ϕ0 − π/2
Thus we should subtract π/2 from the answers in part (b)
23
4-11. Generally, k · r = kx x + k y y + kz z.
(a) For propagation along the z axis, kx = k y = 0, So k · r = kz z, with kz = 2 π/λ. The waveform can then be
written as,

2π
ψ = A sin(k · r − ω t) = A sin(kz z − ω t) = A sin (z − v t) = A sin 2 π(z/λ − v t)
λ
√ 1 2π
(b) In this case, kz = 0 and kx = k y = |k|/ 2 = √ . The general form of the wave is then,
2 λ

2π x y
ψ = A sin(k · r ± ω t) = A sin(kx x + k y y ± ω t) = A sin √ (x + y ± v t) = A sin 2 π √ + √ ±vt .
2λ 2λ 2λ
If one is interested in the wave displacement only on the line x = y,

2x
ψ = A sin 2 π √ ± v t .
2λ
k k
(c) In this case k = √ (x̂ + ŷ + ẑ ) and k · r = √ (x + y + z), with k = 2 π/λ. The waveform is then,
3 3

k 2π
ψ = A sin(k · r ± ω t) = A sin √ (x + y + z) ± ω t = A sin √ (x + y + z ± v t)
3 3λ
4-12. Let z̃ = a + i b where a and b are real.

(a) (z̃ + z̃ *)/2 = (a + i b + a − i b)/2 = a = Re(z̃ )
(b) (z̃ − z̃ *)/2i = (a + i b − a + i b)/2i = b = Im(z̃ )
(c) Let z̃ = eiθ = cos θ + i sin θ and apply the result from (a): cos θ = eiθ + e−iθ /2

(d) Let z̃ = eiθ = cos θ + i sin θ and apply the result from (b): sin θ= eiθ − e−iθ /2i

4-13. (a) Note that eiπ/2 = cos(π/2) + i sin(π/2) = i so that
i A ei(k·r −ωt+ ϕ0) = eiπ/2 A ei(k·r −ωt+ ϕ0) = A ei(k·r −ωt+ ϕ0 +π/2)
(b) Similarly, eiπ = cos(π) + i sin(π) = − 1 so that
− A ei(k·r −ωt+ϕ0) = eiπ A ei(k·r −ωt+ ϕ0) = A ei(k·r −ωt+ ϕ0 +π)
4-14. One can write ψ = Im(ψ̃) with
ψ̃ = A ei(ky+ωt) + A ei(ky −ωt+π) = A ei(ky+ωt) + A eiπei(ky −ωt) = A ei(ky+ωt) − A ei(ky −ωt)

where I have used the result from problem 4 13b. Regrouping gives,
ψ̃ = A eiky eiωt − e−iωt = 2 i A eiky sin(ω t) = 2 A sin(ω t) [i cos(k y) − sin(k y)]

Here I have used the result from problem 4 12d. Reforming the real wavefunction gives the desired relation:
ψ = Im(ψ̃) = 2 A cos(k y) sin(ω t)
4-15. The irradiance is related to the electric field amplitude by the relation Ee = (ε0 c/2)E02 so that,
2000
r r
2 Ee
E0 = = V/m = 870 V/m
ε0c 8.85 × 10−12 × 3 × 108
Then,
B0 = E0/c = 870/(3 × 108) T = 2.90 × 10−6 T
24
4-16. (a) B0 = E0/v = E0 n/c = (100) (1.5)/(3 × 108) T = 5.0 × 10−7 T
(b) hS i = Ee = ε0 c2/2)E0B0 = (8.85 × 10−12 9.0 × 1016 /2) (100) (5.0 × 10−7)W/m2 = 19.9 W/m2
4-17. (a) From the solution to problem 4 15,
2 · 1350
r r
2 Ee
E0 = = V/m = 1010 V/m, B0 = E0/c = 3.37 × 10−6 T
ε0c 8.85 × 10−12 × 3 × 108
Ee Ee 1350
(b) Photon flux = = = m −2 s−1 = 4.76 × 1021 m−2 s−1
h ν h c/λ (6.626 × 10 ) (3 × 108)/(7 × 10−7)
−34
(c) k = 2 π/λ = 2 π/(7 × 10−7 m) = 1.43 × 106/m, ν = c/λ = 4.28 × 1014 Hz. So,
E = (1010 V/m) sin 2 π 1.43 × 106/m s ± 4.28 × 1014 /s t

220 · 0.05
4-18. (a) Ee = Φ/A = W/m2 = 8.75 × 10−3 W/m2,
4 π (10)2
r
2 · 8.75 × 10−3
r
2 Ee
E0 = = V/m = 2.57 V/m
ε0c 8.85 × 10−12 × 3 × 108
2000
(b) Ee = Φ/A = −10 W/m2 = 2 × 1013 W/m2
r10
2 · 2 × 1013
r
2 Ee
E0 = = V/m = 1.23 × 108 V/m, B0 = E0/c = 0.409 T
ε0c 8.85 × 10−12 × 3 × 108
4-19. The energy flow in a cylindrical wave is in the (cylindrical) radial direction. The total flux Φe through a
cylindrical surface of length L and radius r must be independent of the radial coordinate r. Now the total
flux is related to the irradiance Ee by the relation,
Φe = Ee A = Ee 2 π r L
For the flux to be independent of r, the irradiance Ee must be proportional to 1/r. The irradiance is pro
portional
√ to the square of electric field amplitude E0 so the electric field amplitude must be proportional to
1/ r . That is,
1
Ee ∝ ∝ E02 ⇒ E0 ∝ 1/r
p
r
4-20. Eq. (4 44) gives,
λ ′ (1 − v/c)1/2 p
= = (1 − v/c) (1 + v/c)−1
λ (1 + v/c)1/2
A Taylor series expansion indicates that, for small v/c,
(1 + v/c) −1 = 1 − v/c + O(v 2/c2) ≈ 1 − v/c
Using this in the expression for λ ′/λ gives the desired result,
λ′ p
≈ (1 − v/c) (1 − v/c) = 1 − v/c
λ
λ ′ 540
r
1 − v/c
4-21. = = ⇒ v = 0.168 c
λ 640 1 + v/c
λ′
r
1 − v/c
4-22. = 4.8 = ⇒ v = − 0.917 c
λ 1 + v/c
25
4-23. From thermodynamics:
r r
3RT 3 (8.31) (1000)
vrm s = = = 2497 m/s
M 0.004
Since the speed is small compared to c, we can use Eq. (4 45),
λ′
= 1 − v/c ⇒ (λ ′ − λ) = ∆λ = (v/c) λ
λ
2497
706.52 × 10−9 m = 0.0059 nm

∆λ =
3 × 108
Now ∆λ is added to λ0 = λ(v = 0) for receding atoms and subtracted from λ0 for approaching atoms. Thus
an estimate for the Doppler width is
Doppler Width ≈ 2 ∆λ = 0.012 nm
4-24. For the stated conditions the electric fields are,
(a) E = − E y [2 sin(ω t) x̂ + sin(ω t) ŷ] (b) E = − E y [2 sin(ω t) x̂ + sin(ω t − π/2) ŷ ]
(c) E = − E y [2 sin(ω t) x̂ + sin(ω t + π/2) ŷ ] (d) E = − E y [2 sin(ω t − π/4) x̂ + sin(ω t + π/4) ŷ ]
(e) E = − E y [2 sin(ω t) x̂ + sin(ω t + π/4) ŷ ].
I’ll place all the field vectors for each case on the same diagram, showing the evolution over one complete
cycle. For all the curves, the labels indicate the values of ω t. For the linear polarization, the dots mark the
end point of E at the indicated values of ω t
y y y
Linear Elliptical Elliptical
3 π/2 π
0, 2 π 5 π/4
π/4 3 π/4
7 π/4
5 π/4, 7 π/4
π/2 3 π/2 3 π/2
0, π, 2 π x x π/2 x
π/4, 3 π/4 3 π/4 5 π/4 7 π/4
π π/4
π/2 0, 2 π
y y
Elliptical Elliptical
5 π/4
π/4
π/2 0, 2 π π 3 π/2
3 π/4 7 π/4 3 π/4 7 π/4
x x
0, 2 π
π 3 π/2 π/2
5 π/4 π/4
26
Chapter 5 Superposition of Waves
5-1. (a) E1 is a function of u1 = k x − ω t = (3/m) x − (4/s) t. The velocity component of the wave can be deter
mined by setting u1 = constant so that,
du1 dx dx
=0=k − ω ⇒ vx = = ω/k = (4/3) m/s
dt dt dt
Similarly, E2 is a function of u2 = k x + ω t = (3/m) x + (4/s) t and so,
du2 dx dx
=0=k + ω ⇒ vx = = − ω/k = − (4/3) m/s
dt dt dt
The waveforms have the same speed and are oppositely directed.
(b),(c) For notational convenience I will write k = 3/m and ω = 3/s. The requirement is,
5E0 5E0
E1 + E2 = 0 = −
(k x − ω t)2 + 2 (k x + ω t − 6)2 + 2
(k x + ω t − 6)2 + 2 − (k x − ω t)2 − 2 = 0
k x + ω t − 6)2 = (k x − ω t)2
k x + ω t − 6 = ± (k x − ω t)
Choosing the +sign:
k x + ω t − 6 = k x − ω t ⇒ 2 ω t = 6 ⇒ t = 3/ω = 3/(4/s) = 0.75 s
This is the answer to (b). Using the solution with the minus sign,
k x + ω t − 6 = − k x + ω t ⇒ 2 k x = 6 ⇒ x = 3/k = 3(3/m) = 1 m
This is the answer to (c).
5-2. (a) First put the functions in a common form. Since the cosine function is even, I can write,
E1 = 2 cos( − ω t) and E2 = 7 cos(π/4 − ω t)
The associated complex functions are then,
Ẽ1 = 2e−iωt , E2 = 7 eiπ/4e−iωt
Note that E1 = Re(Ẽ1) and E2 = Re(Ẽ2). A phasor diagram displaying the complex amplitudes is shown
below at the left,
(b) Using Eq. (5 9):
p
E0 = E01 2 2
+ E02 + 2 E01 E02 cos(α2 − α1)
p
E0 = 2 + 7 + 2 · 2 · 7 cos(π/4) = 8.53
2 2
E Using Eq. (5 10):

√
E2 E01 sin(α1) + E02 sin(α2) 0 + 7/ 2
7 tan(α) = = √ ⇒ α = 0.619in ≈ 0.2 π
E01 cos(α1) + E02 cos(α2) 2 + 7/ 2
π/4
2 The resultant is then,
E1 E = E + E = E cos(α − ω t) = 8.53 cos(0.2 π − ω t)
1 2 0
5-3. The resultant is, E = 3 cos(π/6 − ω t) + 4 cos(π/2 − ω t). The frequency is ω = 2 π/T = 2 π/s. Using Eqs. (5
9) and (5 10),
p p
2 2
E0 = E01 + E02 + 2 E01 E02 cos(α2 − α1) = 32 + 42 + 2 · 3 · 4 cos(π/2 − π/6) = 6.08
E sin(α1) + E02 sin(α2) 3 sin(π/6) + 4 sin(π/2)
tan α = 01 = ⇒ α = 0.36 π
E01 cos(α1) + E02 cos(α2) 3 cos(π/6) + 4 cos(π/2)
E = 6.08 cos(0.36 π − (2 π/s) t)
27
5-4. One could proceed directly by mimicking the development for the cosine waves leading to Eqs. (5 9) and
(5 10). I choose to first convert the given fields to the cosine form and then using those equations. That is,
y1 = 5 sin(ω t + π/2) = 5 cos(ω t) = 5 cos(0 − ω t)
y2 = 7 sin(ω t + π/3) = 7 cos(ω t + π/3 − π/2) = 7 cos(ω t − π/6) = 7 cos(π/6 − ω t)
Them using Eqs. (5 9) and (5 10):
p
y0 = 52 + 72 + 2 · 5 · 7 cos(π/6) = 11.6
0 + 7 sin(π/6)
tan α = = α = 0.098 π
5 + 7 cos(π/6)
y = 11.6 cos(0.098 π − ω t) = 11.6 cos(ω t − 0.098 π) = 11.6 sin(ω t − 0.098 π + π/2)
y = 11.6 sin(ω t + 0.402 π)
5-5. The frequency of all of the waves is ω = 2 π/T = π/s. Putting all of the waves into cosine form gives,
E1 = cos(π/18 − π/2 − ω t) = cos( − 4 π/9 − ω t)
E2 = 3 cos(5 π/9 − ω t)
E3 = 2 cos(π/6 − π/2 − ω t) = 2 cos( − π/3 − ω t)
Using Eq. (5 13) the amplitude of the resultant wave can be found as
E02 = [sin(4 π/9) + 3 sin(5 π/9) + 2 sin( − π/3)]2 + [cos(4 π/9) + 3 cos(5 π/9) + 2 cos( − π/3)]2
E0 = 0.695
Using Eq. (5 12),
sin( − 4 π/9) + 3 sin(5 π/9) + 2 sin( − π/3)
tan α = ⇒ α = 0.349
cos( − 4 π/9) + 3 cos(5 π/9) + 2 cos( − π/3)
So the resultant wave has the form, E = 0.649 cos(0.349 − (π/s) t)
5-6. (a) If they are in phase the amplitudes add, ER = (100) (0.02 V/m) = 2 V/m
(b) If the sources have random phase differences, the irradiances add. Since the irradiances add, the square
of the amplitudes add. Then
2
ER 2
= (100) E01 ⇒ ER = 10 (0.02 V/m) = 0.2 V/m
5-7. At the indicated position,

ψ1 = A1 cos(8 π/3 − ω t), A1 = 4 cm, ω = 20/s
ψ2 = A2 cos(3 π/2 − ω t), A2 = 2 cm, ω = 20/s
Using Eqs (5 9) and (5 10),
p
A = A21 + A22 + 2 A1 A2 cos(α2 − α1) = 20 + 16 cos(3 π/2 − 8 π/3) cm = 2.48 cm
p
4 sin(8 π/3) + 2 sin(3 π/2)

tan α = ⇒ α = 2.51
4 cos(8 π/3) + 2 sin(3 π/2)
ψR = (2.48 cm) cos(2.51 − (20/s) t)
5-8. Now, v g = v p − λ (dv p/dλ), with v p = c/n.

2πc dv d(c/n) − c dn/n2 ω2 dn
(a) λ = c/ν = . Further, p = = = . Therefore,
ω dλ d(2 π c/ω) − 2 π c dω/ω 2 2 π n2 dω
2 π c ω 2 dn

ω dn
vg = vp − = v p 1−
ω 2 π n2 dω n dω
(b) For normal dispersion, dn/dλ < 0 so that dn/dω > 0 and v g > v p.
28
nD − n C 1
5-9. It is given that ∆ ≡ = . Then,
n0 − 1 30

λ ∆n
v g = v p[1 + (λ/n) dn/dλ] ≈ v p 1 + D
nD ∆λ
1
Here, − ∆n = nC − nF = ∆ (n0 − 1) = (1.50 − 1) = 1/60 and ∆λ = λC − λF = (656.3 − 486.1)nm = 170.2 nm
30
Then,
589

c λD ∆n c 1
vg = 1+ = 1− = c/1.56
nD nD ∆λ 1.50 1.5 60 · 170.2
5-10. Now, n = A + B/λ2 and v g = v p [1 + (λ/n) dn/dλ]. Therefore,

dn 2 (2.5 × 106) −1 −1
= − 2 B/λ3 = − Å = − 4 × 10−5 Å
dλ (5000)3
Then,
5000

c c c c
and v g = − 4 × 10 −5

vp = = 1+ =
n 1.50 1.50 1.5 1.73
√
5-11. (a) We know that K = n2 and v g = v p(1 + (λ/n) dn/dλ). So, v p = c/n = c/ K . Then
√
dn d( K ) 1 K −1/2 dK
= √ = =
n K 2 K 1/2 2K
Also, v = ν λ so that dv = λ dν + νdλ. At a particular velocity, λ dν = − ν dλ, so that

λ ν ω
=− =−
dλ dν dω
Thus,

c dK ω c ω dK
vg = √ 1+ √ − =√ 1−
K K dω K 2K dω
(b) Given K = 1 + A/ ω02 − ω2 where the second term is small compared to the first. Then,

dK − A ( − 2 ω) 2Aω
= 2 2 2
= 2
dω ω0 − ω ) ω0 − ω 2)2
Using the result of (a), ! !
c ω 2Aω c A ω2
vg = √ 1− =√ 1− 2
K 2 K ω02 − ω 2)2 K ω0 − ω 2)2
Now in the second term in the parentheses the difference of K from 1√can be ignored to lowest order. How
ever this difference can not be neglected outright in the pref actor c/ K . In fact, a Taylor series expansion
gives,
1 A
K −1/2 ≈ 1 +
2 ω02 − ω 2

so that keeping only the lowest order corrections,

!
1 A A ω2
vg = c 1 + − 2
2 ω02 − ω 2 ω0 − ω 2)2
Finally as is typically the case of interest, (and if K is close to 1), ω 2 ≫ |ω02 − ω 2| so that the middle term
can be dropped giving, !
A ω2
vg = c 1 − 2
ω0 − ω2)2
29
dω dv
5-12. (a) According to Eq. 5 38, v g = = v p + k p . Then with k = 2 π/λ and dk = − (2 π/λ2)dλ. Thus,
dk dk
2 π d v p λ2 dv
vg = vp + = vp − λ p
λ ( − 2 π) dλ dλ
dv p
(b) v g = v p − λ = (A + B λ) − λ B = A = constant. Since the group velocity is independent of wavelength,
dλ
any waveform constructed from harmonic waves of different wavelengths is transmitted by the medium
without a change in shape.
dv p
5-13. As in problem 5 12a, v g = v p − λ . Then,
dλ
Long λ:
dv p d g 1 g 1
q q
= = = vp
dλ dλ 2πλ 2 2πλ 2
Short λ:
1/2 1/2
dv p d 2πT 1 2πT
= =−
dλ dλ λ ρ 2λ λ ρ
So,
1/2 1/2 1/2
2πT 1 2πT 3 2πT 3
vg = + = = vp
λρ 2 λρ 2 λρ 2
5-14. The frequency received by the mirror is ν1′ = ν0/(1 + v/c). Since λ ′/λ = ν/ν ′ = 1 − v/c and v < 0 due to
recession. The reflected beam is also due to a receding source so the reflected frequency is
ν1′
ν2′ =
1 + v/c
Thus,
ν0 1 ν0
ν2′ = = = ν0(1 + v/c) −2
1 + v/c 1 + v/c (1 + v/c)2
For v/c ≪ 1, a Taylor series expansion is appropriate, so that ν2′ ≈ ν0 (1 − 2v/c) and the beat frequency is
ν2′ − ν0 = 2(v/c) ν0
5-15. If the wave is inverted upon reflection, the standing wave will be of the form given in Eq. (5 23),
yS = 2 y0 sin(k x) cos(ω t)
From the given traveling wave information, k = 4/cm, ω = 2 π/T , and y0 = 7 cm. So, the standing wave has
the form,
y = (14 cm) sin[(4/m) x] cos[(2 π/T ) t]
This wave has an amplitude of 14 cm, a wavelength of λ = 2 π/k = 1.57 cm, in the sense that it is a standing
wave, the net speed of energy transmission is zero. The length of one loop is λ/2 = 0.785 cm. The period is
T.
30
5-16. The standing wave has the form,
π x 50 π
y = 2 A sin(k x) cos(ω t) = (3 cm) sin cos t
10 cm s
(a) The amplitude of each of the oppositely directed traveling waves that add to the standing wave is
A = 1.5 cm. One of these component waves is traveling in the + x direction and the other is traveling in the
− x direction. The wave speed of the component waves is v = ω/k = (50 π/(π/10)) cm/s = 500 cm/s. The
wavelength is λ = 2 π/k = (2 π)/(π/10)cm = 20 m. The frequency is ν = ω/2 π = 25 Hz.
(b) The internodal distance is λ/2 = 10 cm
(c) The displacement, velocity component, and acceleration component at x = 5 cm and t = 0.22 s are,
y(x = 5 cm, t = 0.22 s) = − 3 cm
dy
vy = = − 2 A ω sin(k x) sin(ω t)
dt
v y(x = 5 cm, t = 0.22 s) = 0
dv y
ay = = − ω 2 2 A sin(k x) cos(ω t) = − ω2 y
dt
a y(x = 5 cm, t = 0.22 s) = 7.40 × 104 cm/s2
5-17. In the complex representation,

h i
E1 = E0 sin(ω t + k x) = Im E0 ei(ωt+kx)
h i
E2 = E0 sin(ω t − k x − ϕR) = Im E0ei(ωt−kx− ϕR)
h i
ER = E1 + E2 = E0 Im ei(ωt+kx) + ei(ωt−kx−ϕR)
h i
ER = E0 Im ei(ωt− ϕR/2)ei(kx+ ϕR/2) + ei(ωt− ϕR/2)e−i(kx+ ϕR/2)
n h io
ER = E0 Im ei(ωt− ϕR/2) ei(kx+ ϕR/2) + e−i(kx+ϕR/2)
h i
ER = E0 Im ei(ωt− ϕR/2) 2 cos(k x + ϕR/2)
h i
ER = 2 E0cos(k x + ϕR/2) Im ei(ωt− ϕR/2)
ER = 2 E0cos(k x + ϕR/2) sin(ω t − ϕR/2)
5-18. The separation between cavity modes is ∆ν = νm+1 − νm = (m + 1) c/2d − m c/2d = c/2d. That is,
∆ν = 3 × 108 m/s /(2 · 1 m = 0.15 GHz

Then the number of lasing modes would be the ration of the frequency range for lasing to the separation of
cavity modes:
6 GHz
#of lasing modes = = 40
0.15 GHz
31
Chapter 6 Properties of Lasers
6-1. The allowed energies of the electron in the hydrogen atom are En = − (13.6 eV)/n2, n = 1, 2, 3, ... The
energies required for this problem are:
E1 = − 13.6 eV E2 = − 3.40 eV E3 = − 1.51 eV E4 = − 0.850 eV E5 = − 0.544 eV
The wavelength associated with a transition is
hc 1240 eV · nm
λmn = =
Em − En Em − En
(a) Lyman Series:
1240 1240
λ21 = nm = 122 nm λ31 = nm = 103 nm λ41 = 97.3 nm
− 3.4 − ( − 13.6) − 1.51 − ( − 13.6)
These are in the ultraviolet (shorter than visible).
(b) Balmer Series:
1240 1240
λ32 = nm = 656 nm λ42 = nm = 486 nm λ52 = 434 nm
− 1.51 − ( − 3.4) − 0.85 − ( − 3.4)
These are in the visible range.
6.2. (a) From the solution to problem 6 1 it is evident that an energy of 5 eV does not correspond to the energy
difference between the ground state of hydrogen and an excited state of hydrogen. Therefore, since the
photon must give up its energy in one lump, the photon will not be absorbed by the hydrogen atom.
(Neglecting very unlikely nonlinear processes, of course.)
(b) An electron in hydrogen is ionized (leaves the proton) when its energy exceeds zero. Thus the energy of
the photon must exceed Efinal − Einitial = 0 − ( − 13.6 eV) = 13.6 eV. Any energy greater than this is possible
since the free electron can have any energy. The corresponding wavelength range is,
hc 1240 eV · nm
λ< = = 91.2 nm
13.6 eV 13.6 eV
(c). In this case the photon energy must exceed 0 − E2 = 0 − ( − 3.4 eV) = 3.4 eV. The wavelength range is
then,
hc 1240 eV · nm
λ< = = 365 nm
3.4eV 3.4 eV
6-3. (a) The rotational inertia is
I = 2 mH r 2 = 2(1.67 × 10−27 0.074 × 10−9/2)2 kg · m2 = 4.6 × 10−48 kg · m2

(b) ∆E = E1rot − E0rot = ~2/I = (6.626 × 10−34 /2 π)2/(4.6 × 10−48 ) J = 2.4 × 10−21 J = 0.015 eV
8.62 ×10 −5 ·293)

(c) P1/P0 = e−(∆E)/kB T = e− 0.015/ = 0.55
6-4. (a) ∆E = E1vib − E0vib = h f = (6.626 × 10−34 ) 1.3 × 1014 J = 8.6 × 10−20 J = 0.54 eV

8.62 ×10 −5 ·293)

(b) P1/P0 = P1/P0 = e−(∆E)/kB T = e−0.54 / = e−21.4 ≈ 5 × 10−10
32
6-5. The first few states of the energy level diagram in the electronic ground state is shown below. The energy
labels are of the form Ek,l. Associated with each vibrational state there are a number of rotational sub
levels. An expanded view of each vibrational state with associated sublevels is shown at the right.
E2,0 + 0.06 eV
E2,0 + 0.015 eV
E2,0
E2,0 = E0,0 + 1.08 eV

E1,0 + 0.06 eV
E1,0 + 0.015 eV
E1,0 = E0,0 + 0.54 eV
E1,0
E0,0
E0,0 + 0.06 eV
E1,0 + 0.015 eV
E1,0
6-6. The energy separation of the states is ∆E = h c/λ = (1240 eV · nm)/(632.8 nm) = 01.96 eV
8.62 ×10 −5 ·300)
Neglecting degeneracies:N2/N1 = P2/P1 = e−(E2 −E1)/kBT = e−1.96 / ≈ e−76 .
6-7. The differential equation can be integrated as shown below,

dN2 dN2
= − A21 N2 ⇒ = − A21 dt
dt N2
Z N2(t) Z t
dN2
= − A21 dt ′ ⇒ ln(N2(t)/N20 ) = − A21 (t − 0) = − A21 t
N2 0 N2 0
N2(t)
= e−A2 1 t ⇒ N2(t) = N20 e−A2 1 t
N20
At time t = 1/A21 , N2(t) = N20 e−A2 1 (1/A2 1 ) = N20 e−1 = N20 /e
2 π h c2 1
6-8. The black body spectral radiance is Mλ = .
λ5 ehc/λkBT − 1
This function is maximized when it’s denominator is minimized:

d h 5 hc/λkBT i
5 hc/λkBT h c 1 hc/λkBT

4
λ (e − 1) = 0 ⇒ λ e − 2 + e −1 5λ =0
dλ kB T λ
Let x = h c/(λ kB T ): λ4 [ − x ex + 5 (ex − 1)] = 0 ⇒ x ex = 5 ex − 1. This is a transcendental equation in x

with solution, x = 4.999986. Then,
hc hc
xm ax = 4.999986 = ⇒ λm ax T = = 2.898 × 10−3 m · K = 2898 µm · K
λm ax kB T 4.999986 kB
6-9. Let x = h c/(λ kB T ). Then dx = (h c/kB T ) − dλ/λ2 so that, dλ = − λ2 (kB T /h c) dx. Then,

∞ ∞
2 π h c2 2 π (kB T )4 ∞
x3 2 π (kB T )4 π 4 4 5
Z Z Z
1 2 kB π
M= Mλ dλ = dλ = = = T4
0 0 λ5 ehc/λkBT − 1 h 3 c2 0 ex − 1 h 3 c2 15 15 h3 c2
4 5

2 kB π
T 4 ≡ σ T 4 = 5.668 × 10−8 W/m2 K4

M=
15 h3 c2
33
6-10. As in the solution to problem (6 9),
2 π c2 1 hc
Mλ = , with x =
λ5 ex − 1 λ kB T
Then,
Z ∞
M= Mλ dλ
0
The change of variable is accomplished by noting that for λ = 0, x = ∞ and for λ = ∞, x = 0. Further,
λ2 kB T

hc dλ
dx = − 2 ⇒ dλ = − dx
kB T λ hc
So,
Z 0 Z ∞ Z ∞ 3
1 1 − λ2 kB T 1 dx x kB T dx
M = 2 π h c2 5 x
dx = 2 π c k B T = 2 π c kB T
∞ λ e −1 hc 0 λ 3 ex − 1 0 hc ex − 1
2 π kB T )4 ∞ x3
Z
M= dx
h 3 c2 0 ex − 1
The definite integral has the value:
∞
x3 π4
Z
dx =
0 ex − 1 15
so that,
4 5
4
W

2kB π
M= T 4 ≡ σ T 4, σ = 5.668 × 10−8
15 h3 c2 m2 K4
6-11. The original temperature is determined from the Wien displacement law,
2898 µm · K 2898 µm · K
λmax T1 = 2898 µm · K ⇒ T1 = = = 5269 K
λmax 0.55 µm
For the total radiant exitance to double,
M2 = 2 M1 ⇒ σ(T2)4 = 2σ T14 ⇒ T2 = 21/4 T1 = 6266 K
The new maximum wavelength is,
2898 µm · K 2898 µm · K
λm ax = = = 0.462 µm = 462nm
T2 6266 K
6-12. The gray body at T2 must match the the black body at the lower temperature, T1. That is,

(MGB )T2 = (MBB )T1 ⇒ (εMB B = (MB B )T1
T2
ε σ T24 = σ T14 ⇒ T24 = T14/ε = (5000 K)4/0.45 ⇒ T2 = 6105 K
2 π h c2 1
6-13. On the absolute temperature scale 451◦F = 506 K. Then Mλ = ε 5 hc/λk
. The plot is as shown.
λ BT
e −1
2e+08
1.6e+08
Mλ (W/m3)
1.2e+08
8e+07
4e+07
0
0 2 4 6 8 10 12 14
λ (µm)
34
6-14. Equation (6 12) indicates that A21 /B21 is inversely proportional to the cube of the wavelength. This ratio
is the ratio of spontaneous emission to stimulated emission. Stimulated emission contributes to the laser
gain while spontaneous emission (essentially) does not. Thus, all else being equal, the gain on a transition
resonant with short wavelength radiation is less than that of a transition resonant with long wavelength
radiation.
6-15. The frequency bandwidth ∆ν = ν2 − ν1 is related to the wavelength range ∆λ = λ2 − λ1 by,
c c λ − λ2 ∆λ λ2|∆ν |
∆ν = − =c 1 ≈ − c 2 ⇒ |∆λ| =
λ2 λ1 λ2 λ1 λ c
2
1.06 × 10−6 1.2 · 1011

From Table 6 1, λ = 1.06 × 10−6 m, so, |∆λ| = m = 4.5 × 10−10 m = 0.45 nm
(3 · 108)
6-16. (a) tc ≈ 1/∆ν = 1/(105 Hz) = 10−5 s Lc = c tc = 3 · 108 m/s 10−5 s = 3000 m

(b) tc ≈ 1/∆ν = 1/(2 · 109 Hz) = 5 · 10−10 s Lc = c tc = 3 · 108 m/s 10−5 s = 0.15 m = 15 cm

6-17. The separation between cavity modes is ∆ν = νm+1 − νm = (m + 1) c/2d − m c/2d = c/2d. That is,
∆ν = 3 × 108 m/s /(2 · 0.5 m = 0.3 GHz

Then the number of lasing modes would be the ration of the frequency range for lasing to the separation of
cavity modes:
2 GHz
#of lasing modes = = 6.67 ⇒ 6
0.3 GHz
6-18. (a) The half angle beam spread is,
λ 6.33 · 10−7 m
θ= = = 4.0 × 10−4 rad = 0.023◦
π w0 π · 5 · 10−4 m
(b) The diameter is about,
D = 2 d θ = 2 (1000 m) 4.0 × 10−4 = 0.8 m = 80 cm

6-19. Assuming that that irradiance is constant over the spot:
P P
Ilaser = laser = laser = Isun
Alaser πD 2/4
r
4 · 0.01
r
4 Plaser
D= = m = 0.0036 m = 3.6mm
π Isun π 1000
35
6-20. (a) The energy difference ∆E between the levels must have near the same energy as the pump photons.
That is,
hc hc 1240 eV · nm
∆E = h ν =
⇒λ= =
λ ∆E ∆E
Then, λ1.53 = (1240/1.53) nm = 810 nm. Similarly, λ1.653 = 750 nm, λ2.119 = 585 nm, and λ2.361 = 525 nm.
(b) The quantum efficiency η Q is,
Energy of an output photon h c/λout λ
ηQ = = = p
Energy of a pump event h c/λ p λout
So using the results from part (a) for the various transitions:
810 750 585 525
η Q,1.53 = = 0.76, η Q,1.653 = = 0.70, η Q,2.119 = = 0.55, and η Q,2.361 = = 0.49.
1064 1064 1064 1064
6-21. (a) The laser output beam would have a power of Poutput = 0.8 × 0.3 × 0.7 × 0.15 × 0.5 × 2500 W = 31.5 W.
Poutput 31.5
(b) The overall efficiency is η = = = 0.0126.
Ppum p 2500
6-22. The diode laser beam is generated in a very small region. The radius w0 of the beam waist of the diode
laser beam must therefore by very small (compared to other types of lasers). As indicated by Equation (6
16), the divergence angle is inversely proportional proportional to w0 and so will be larger for diode lasers
than for other types of lasers.
6-23. As the irradiance in the laser cavity increases, the population inversion decreases. Given that there is a
positive population inversion, the population densities of the upper and lower lasing levels satisfy the
inequality Nupp er > Nlower . The magnitude of the net rate of depletion of the population inversion due to
stimulated processes is proportional to I (Nupp er − Nlower ). Thus increased irradiance causes a larger rate of
depletion of the population inversion, leading to a lessened inversion.
6-24. The diode laser frequency can be tuned to match one of the pump transitions (see problem 6 20) of the
Nd:YAG gain medium. Thus nearly all of the diode laser pump energy can be used to pump the laser.
Much of the arc lamp energy (see Figures (6 8) and (6 9)) is not resonant with a pump transition and so is
wasted. This wasted pump energy lowers the overall efficiency of the laser system.
36
Chapter 7 Interference of Light
2
7-1. (a) I1 = (ε0 c/2) E01 = 8.85 × 10−12 · 3 × 108/2 3000 )2 W/m2 = (0.0013275) 3000)2 = 11, 950 W/m2

2
I2 = (ε0 c/2) E02 = (0.0013275) 4000)2 = 21, 240 W/m2
√ √
(b) I12 = 2 I1 I2 cos δ = 2 I1 I2 cos [π/3 − (π/5 − π/6)] = 18, 720 W/m2
(c) I = I1 + I2 + I12 = 51, 900 W/m2
(d) The visibility is,

√ √ √ √
Imax − Imin I1 + I2 + 2 I1 I2 − (I1 + I2 − 2 I1 I2 ) 4 I1 I2 2 I1 I2
visibility = = √ √ = = = 0.96
Imax + Imin I1 + I2 + 2 I1 I2 + I1 + I2 − 2 I1 I2 ) 2 (I1 + I2) I1 + I2
7-2. As in the solution to problem 7 1,

√
2 I1 I2
visibility =
I1 + I2
Since irradiance is proportional to the square of the field amplitude, for parallel fields, this relation can be
rewritten as,
2E E 2 (1.6) (2.8)
visibility = 2 01 01
2 = 1.62 + 1.82 = 0.86
E01 + E02
For perpendicular fields, I12 ∝ E1 · E2 = 0 so the visibility is zero.
7-3. As in the solution to problem 7 2,

√
2 I1 I2 2E E
visibility ≡ V = = 2 01 02
2
I1 + I2 E01 + E02
Then,
2
2 E01 E02 1/E02 2R
V= 2 2 2 = R2 + 1 , with R = E01 /E02
E01 + E02 1/E02
For R = 2,
V = 0.8
If V = 0.5,
V (1 + R2) = 2 R ⇒ 0.5 + 0.5 R2 = 2 R
√
4± 16 − 4 √
R2 − 4 R + 1 = 0 ⇒ R = = 2 ± 3 = 3.732, 1/3.732
2
√ √ √ √
Im ax − Im in I1 + I2 + 2 I1 I2 − (I1 + I2 − 2 I1 I2 ) 4 I1 I2 2 I1 I2
7-4. (a) visibility = = √ √ = =
Im ax + Im in I1 + I2 + 2 I1 I2 + I1 + I2 − 2 I1 I2 ) 2 (I1 + I2) I1 + I2
So if I1 = N I2,
√ √
2 N I2 2 N
visibility ≡ V = =
(N + 1) I2 N + 1
(b) Solving the above relation for N ,
"√ #
1 − V2 + 1
N=
V
For V = 0.96, N = 1.78. For V = 0.90, N = 2.55. For V = 0.8, N = 4. For V = 0.5, N = 13.9
37
λL
7-5. See Figure 7 24 in the test. Using m λ = a sin θ = a y/L leads to ∆y ≡ ym+1 − ym = . So the fringe sepa
m
ration is,
546.1 × 10−7 · 100
∆y = cm = 0.0273 cm
0.2
The irradiance on the screen relative to y = 0 at a maximum is given by,
π 0.2

2 πay
h i
I = 4 I0 cos = 4 I 0 cos 2
y = 4 I0 cos2[(115/cm)y]
λL 546.1 × 10−7 · 100
To take into account phase changes on reflection, and using y = 0 to be the plane of the mirror surface
where I = 0, one can use the sin function instead,
I = 4 I0 sin2[(115/cm)y]
7-6. Let λ1 = 436 nm and the unknown wavelength be λ2. At the position of overlap,
(3 + 1/2) λ1 = a sin θ = 3 λ2 ⇒ λ2 = (3.5/3) λ1 = (7/6) (436 nm) = 508.7 nm
7-7. See Figure 7 25 in the text. The position of the mth minima relative to the central maximum is given by,
(m − 1/2) λ = a sin θ = a ym/L ⇒ ym = L (m − 1/2)λ/a
The distance on the screen between the fifth minima on either side of the central maximum is
∆y ≡ y5 − y−5 = [4.5 − ( − 4.5)] sLλ/a = 9 L λ/a
∆y a (3.473) (0.02 )
λ= = cm = 5.15 × 10−5 cm = 514 nm
9L 9 (150)
7-8. The set up is like that of Figure 7 3 in the text with L = 7 m and a = 1 mm. The single slit is used to
render the the quasi monochromatic light source more coherent. If a laser is used, the single silt is not
needed. The wavelength of the light can be determined using Eq. (7 23),
∆y a (0.0056) (0.001)
ym+1 − ym ≡ ∆y = λ L/a = λ = = m = 8.0 × 10−7 m = 800 nm
L 7
7-9. The fringe separation is,

ym+1 − ym ≡ ∆y = λ L/a
(a) So the slit to screen distance should be,
∆y a (0.001) (0.0005)
L= = m = 0.833 m
λ 6 × 10−7
(b) The optical path difference can be written in terms of wavelengths as ∆ = m λ. The with and without
the plate of thickness t,
nt−t t 10−4
∆2 − ∆1 = ∆m λ ⇒ ∆m = = (n − 1) = (1.5 − 1) = 83.3 fringes
λ λ 6 × 10−7
π∆
(c) I = 4 I0 cos2 where ∆ = a y/L. At ∆ = 0, I = Im ax = 4 I0. Then for I = 2 I0 = Imax /2:
λ
π∆
2 I0 = I = 4 I0 cos2 ⇒ ∆ = λ/4 = 150 nm
λ
7-10. At a pinhole, destructive interference occurs for wavelengths given by (m + 1/2) λ = a sin θ = a y/L. So,
y (1.25) (3) 2500
λ= = mm = nm
(m + 1/2)L (m + 1/2) 1500 m + 1/2
In the visible range, m = 4 and m = 5 give the wavelengths 556 nm and 455 nm.
38
7-11. See Figure 7 26 in the text. Constructive interference occurs at screen locations,
m λ(d + L) m λ(d + 2 d) m λ(3 d)
y= = =
a a 2 d α (n − 1)
3 λ ∆m 3 589.3 × 10−9 1
α= = = 0.005893 rad = 0.3376◦ = 20.3 ′
2 n − 1 ∆λ 2 1.5 − 1 3 × 10−4
7-12. See Figure 7 27 that accompanies the statement of the problem in the text. Also refer to the discussion
related to Figure 7 8 in the body of the text.
m λ s m λ (d + D)
ym = =
2θd 2θd
So,
λ (d + D)
ym+1 − ym ≡ ∆y =
2θd
λ (d + D) (589.3 × 10−7) (50 + 100)
θ= = = 0.001768 rad = 6 ′5 ′′
2 d∆y 2 · 50 · 0.05
7-13. Refer to Figure 7 9 for the setup. Equation (7 25) gives, a = 2 d α (n − 1). Using this in Eq. (7 22) gives,
λL λL λ (d + s)
∆y = = =
a 2 d α (n − 1) 2 d α (n − 1)
Rearrangement leads to,
λ (d + s)
α=
2 n (n − 1) ∆y
It is given that ∆y = 0.5 cm/19. Then,
546.1 × 10−7) (d + 4 d)
α= = 0.010376 rad = 35 ′40 ′′
2 d (1.5 − 1) 0.5/19
7-14. The condition for a minimum in the reflecting light is

∆ p + ∆r = = 2 n t = (m + 1/2) λ
For the two wavelengths then
2 n t = (m1 + 1/2) λ1 = (m2 + 1/2) λ2
m1 + 1/2 λ2 675
= = = 1.2875
m2 + 1/2 λ1 525
By trial and error, this relation is satisfied with m1 = 4 and m2 = 3. Then,
(m1 + 1/2) λ1 (4.5) (525 nm)
t= = = 908.65 nm
2n 2 (1.30)
7-15. At normal incidence, (m + 1/2) λ = 2 n t. At 45◦, (m + 1/2) λ ′ = 2 n t cos θt.

Here θt is the angle the ray makes with the normal in the film which can be found from Snell’s law,
1 1
sin(45◦) = n sin θt ⇒ sin θt = √ =√ = 0.5124 ⇒ θt = 30.825◦ = cos θt = 0.859
2n 2 (1.38)
Then,
2 n t cos θt (m + 1/2) λ cos θt
λ′ = = = λ cos θ1 = (580 nm) (0.859) = 498 nm
m + 1/2 m + 1/2
√
7-16. Using Eq. (7 28),the index of refraction of the film is n f = n0 ns = (1) (1.78) = 1.33.
p
The required thickness is t = λ f /4 = λ/(4 n f ) = (550 nm)/(4 · 1.33) = 103 nm
39
1−n 1 − 1.4
7-17. (a) r = where n = n2/n1. So r = = − 1/6 and R = r 2 = 0.0278. That is 2.78% of the light is
1+n 1 + 1.4
reflected.
(b) The thickness should be t = λnf /4 = (500 nm)/(4 · 1.4) = 89.3 nm
(c) The reflection from the bottom of the film layer is
1 − 1.6/1.4
rb ottom = = − 1/15
1 + 1.6/1.4
An estimate gives
2
reffective = (|rtop | − |rb ottom |)2 = (1/6 − 1/15)2 = 0.01 ⇒ 1% energy reflected
The reflection coefficients from the top and bottom are subtracted because the film thickness is chosen for
destructive interference so the emerging reflected beams are π out of phase. This estimate ignores the inte
rior reflection at the top surface as the beam reflected from the bottom surface emerges from the film. The
light lost at this interior reflection is a small part of the total.
7-18. The soap film forms a wedge like structure since the soapy water collects at the bottom of the wire frame.
This structure is shown in the diagram below.
We assume a wedge angle θ. Interference occurs by reflection of the incident
m=0
632.8 nm light. Consider dark fringes: m λ = 2 n t. At m = 0, ∆ = λ/2 due to
632 nm 1 cm
the relative phase shift on reflection. At the film bottom,where m = 15,
θ
m = 15 (15) λ = 2 (1.33) t = 2.66 (1 cm) θ ⇒ θ ≈ 15 λ/2.66 = 3.57 × 10−4 rad = 1 ′14 ′′

t
7-19. See Figure 7 28 that accompanies this problem in the text. The dark lines are wavelengths for which
destructive interference occurs on reflection. These satisfy.
m λ = 2 n t cos θt
Here the film is the air layer. The angle in the air film is the same as the incident angle of 45◦ since the
angle that the ray emerges from the top glass slide into the air film is the same as the angle at which the
ray entered the glass slide from the top ambient air. Then,
2 n t cos (45◦) 2 · 1 · (104 nm) 0.707 14, 142 nm
λm = = =
m m m
There are 15 orders in the visible with m ranging from 21 to 35. The dark lines occur at:
λ21 = 673.4 nm, λ22 = 642.8 nm..., λ35 = 404.1 nm.
7-20. Refer to Figure 7 15b in the body of the text. Constructive interference will occur for
(m + 1/2) λ = 2 t
Here, t is the thickness of the air wedge at a given horizontal position. The 40 th bright fringe corresponds
to m = 39 since the first bright fringe occurs for m = 0. Then, for the 40th fringe,
(39.5)(589 × 10−7)
39.5 λ = 2 t ⇒ t = cm = 1.16 × 10−3 cm
2
7-21. Refer to the figure below. The maximum thickness of the air wedge is d = 0.05 mm and the length of the
wedge is 20 cm. Let the horizontal distance from the point of contact of the glass plates to a given position
in the air wedge be x. For dark fringes, 2 t + λ/2 = (m + 1/2) λ ⇒ t = m λ/2
mλ
From the geometry: x/t = (0.2 m)/d so x = (t/d) (0.2 m) = (0.2 m).
2
(0.1 m) 546.1 × 10−9 m)
For ∆m = 1, ∆x = (0.1 m) λ/d = = 1.09 mm
d
5 × 10−5 m
xd (0.2 m) (5 × 10 m)
−5
x
Also, solving for m = = = 181.3
2 0 cm (0.1 m) λ (0.1 m) (546.1 × 10−9 m)
Counting the m = 0 dark fringe, 184 fringes would appear.
40
7-22. The physical arrangement is sketched below.
The two image positions viewed by the reflected light are
I1 and I2′′′. These serve as the virtual sources. There position
I1′
can be determined in a step by step fashion. The point source
is O. The virtual source I1 due to the top surface reflection is
O located a distance s below the top surface. The position of
s
the virtual source I2′′′ is found as:
Reflection at top surface forming I2′ :
s 1 1
+ = 0 ⇒ s2′ = − n s
I1
s s2′
Reflection at bottom forming I2′′:
n
2 t/n
s2 = t − s2′ = t + n s ⇒ s2′′ = − s2 = − (t + n s)
I2′′′
Refraction out of top surface forming I2′′′:
s2 = t − s2′′ = 2 t + n s

n 1 ′′′ s2 2t
+ = 0 ⇒ s2 = − = − s +
s2 s2′′′ n n
I2′′
That is I2′′′ is formed a distance of s + 2 t/n below the top
surface.
Thus the separation of virtual sources is 2 t/n.
7-23. Refer to Figure 7 17 in the text. Using Eq. (7 38),

2 tm + ∆r = 2 tm + λ/2 = m λ ⇒ tm = (m − 1/2) λ/2
The 10th bright fringe occurs for m = 10, so that
t10 = 9.5 λ/2 = (9.5 /2) (546.1 × 10−6 mm) = 2.59 × 10−3 mm
Using Eq. (7 39), the radius of curvature R of the lens surface can be found:
2
r10 + t210 7.89/2)2 + 2.59 × 10−3)2
R= = = 3000 mm = 3m
2 t10 2 (2.59 × 10−3)
7-24. Using Eq. (7 39) with tm ≪ rm,

2
rm,air ≈ 2 R tm,air = 2 R (m − 1/2) λ/2
2
rm,liq ≈ 2 R tm,liq = 2 R (m − 1/2) λ/2nliq
2
rm,air r √
2 = nliq ⇒ m,air = nliq
rm,liq r m,liq
7-25. Using Eq. (7 39) with tm ≪ rm,

2 2
r11,1 ≈ 2 R t11,1 = 2 R (11 − 1/2) λ1/2 = r10,2 ≈ 2 R t10,1 = 2 R (10 − 1/2) λ2/2
λ2 = λ1 (10.5/9.5) = (546 nm)(10.5/9.5) = 603.5 nm
p
r11 = 2 R (11 − 1/2) λ1/2 = 2 (1) (10.5) (546 × 10−9)/2 m = 0.002398 m = 2.39 mm
p
t11 = 10.5 λ1/2 = 2.87 × 10−4 cm
7-26. Refer to Figure 7 20 and the surrounding discussion in the body of the text.
546.1

∆m = ∆x/x = = 3.4/1 = 3.4 ⇒ d = ∆m λ/2 = (3.4) nm = 928.4 nm
2
41
7-27. Refer to Figure 7 30 that accompanies the statement of the problem in the text. The cross sections of the
emergent reflected and transmitted beams are the same as that of the incident beam. Inside the film the
cross section is modified due to refraction.
Φ Φ 10−3 W
(a) Ee = = 2= = 1273 W/m2 = (ε0 c/2) E02 ⇒ E0 = 980 V/m.
A πr π 5 × 10−4)2 m2
(b) Snell’s law gives: (1) sin(45◦) = (1.414) sin θf ⇒ θ f = 30◦
(c) Using the Stokes relations,

|r ′| = |r| = 0.28
t t ′ = 1 − r 2 = 0.9216
(d) Reflected beams:
E1 = r E0 = (0.28) 980 V/m) = 274 V/m, (E1/E0)2 = 0.078 = 7.8%
E2 = r ′ t t ′ E0 = (0.28) (0.9216) (980 V/m) = 253 V/m, (E2/E0)2 = 0.067 = 6.7%
E3 = (r ′)3t t ′ E0 = 0.28)3 (0.9216) (980 V/m) = 19.8 V/m, (E3/E0)2 = 0.00041 = 0.041%
(e) Transmitted beams:
E1 = t t ′ E0 = (0.9216) (980 V/m) = 274 V/m, (E1/E0)2 = 0.85 = 85%
E2 = (r ′)2 t t ′ E0 = (0.28)2(0.9216) (980 V/m) = 70.8 V/m, (E2/E0)2 = 0.0052 = 0.52%
λ 632.8 nm
(f) m λ = 2 n f t cos θ f ⇒ t = = = 258 nm
2 n f cos θ f 2 (1.414) cos(30◦)
7-28. From Eq. (7 27) and the Stokes relations,

1 − n 1 − 1.52/1

′
|r| = |r | =
= = 0.206
1 + n 1 + 1.52/1
t t ′ = 1 − r 2 = 0.95742
(a) Reflected beams:
E1 = r E0 = (0.20635) (1 V/m) = 0.206 V/m
E2 = r ′ t t ′ E0 = (0.20635) (0.95742) (1 V/m) = 0.198 V/m
E3 = (r ′)3 t t ′ E0 = (0.20635)3 (0.95742) 1 V/m) = 0.084 V/m
Transmitted beams:
E1 = t t ′ E0 = (0.95742) (1 V/m) = 0.957 V/m
E2 = (r ) t t E0 = (0.20635)2 (0.95742) 1 V/m) = 0.041 V/m
′ 2 ′
E3 = (r ′)4t t ′ E0 = (0.20635)4 (0.95742) 1 V/m) = 0.0017 V/m

2R
(b) As shown in problem 7 3 the visibility is V = , where R is the ratio of the amplitudes.
R2 + 1
2 (1.044)
Reflected: R = E1/E2 = 0.206/0.198 = 1.044. V = = 0.999
1 + 1.0442
2 (23.485)
Transmitted: R = E1/E2 = 0.957/0.041 = 23.485. V = = 0.085
1 + 23.4852
42
Chapter 8 Optical Interferometry
8-1. λ = (2 ∆d/∆m) = (2 · 0.014 cm)/523 = 4.6 × 10−5 cm = 436 nm
8-2. Straight fringes are due to a wedge between one mirror and the image of the other (M 2 and M 1 ′ in Figure
8 1). Interference then occurs as from reflection by an air wedge.
5 4 6 .1 n m There are 12 fringes/cm so there are 11 fringe spaces/cm.
θ d m λ = 2 d ⇒ d = m λ/2 = (11/2) λ = 5.5 λ.
1 cm θ = t/(1 cm) = 5.5 5.461 × 10−6)/1 = 3.00 × 10−4 cm = 0.0172◦ = 1 ′2 ′′
8-3. The optical path difference due to the insertion of the thin sheet of width t and index n is
∆ = m λ = 2 (n t − t) = 2 t (n − 1)
mλ 35 (589 × 10−9 m)
t= = = 23.75 × 10−6 m = 23.75 µm
2 (n − 1) 2 (1.434 − 1)
8-4. (a) Using Eq. (8 5), mmax = 2 d/λ = (2 · 2 cm)/(500 × 10−7 cm) = 80, 000. (b) m = mm ax − 6 = 79, 994
8-5. (a) The optical path length due to the presence of the cell of length L and index of refraction n is
∆ = N λ = 2 n L − 2 L = 2 L(n − 1) ⇒ n = 1 + N λ/(2 L)
2 (0.1 m)(1.00045 − 1)
(b) Rearranging the result from (a), N = 2 L(n − 1)/λ = = 153
589 × 10−9 m
8-6. At θ = 0, m = 2 d/λ = (20 µm)/(0.6238 µm) = 31.6. The smallest diameter dark ring corresponds to m = 31.
Then,
31 λ 31 (0.6328)
m λ = 2 d cos θm ⇒ cos θ31 = = = 0.98084 ⇒ θ31 = 11.23◦
2d 20
The 10th dark ring is then of order m = 22:
22 λ 22 (0.6328)
cos θ22 = = = 0.69608 ⇒ θ22 = 45.89◦
2d 20
8-7. In general after a path length difference of d, m λ = 2 d. A defect of depth ∆d then satisfies 2 ∆d = λ ∆m.
So, ∆d = ∆m λ/2 = (1/4) (632.8 nm)/2 = 79.1 nm or λ/8.
8-8. (a) The resolving power is,

λ 656.3
R= = = 48, 257 ≈ 48, 260
∆λ 1.360
(b) The plate thickness d is d = m λ/2. The mode number m can be found by comparing the coefficient of
finesse, F and the resolving power. The coefficient of finesse is,
4 r2 4 (0.99)2
F= = = 9900
(1 − r 2)2 (1 − (0.99)2)2
The resolving power is
π √ 2R (2) (48, 260)
R= m F ⇒m= √ = √ = 308.8 ≈ 309
2 π F π 9900
Then
λ 656.3 nm
d=m = (309) = 0.01013 cm
2 2
43
8-9. (a) The coefficient of finesse is,
4 r2 4 (0.999)
F= = = 3.996 × 106
(1 − r 2)2 (1 − 0.999)2
(b) The frequency separation between the laser modes is related to the wavelength separation,
c λ2 (632.8 × 10−9) (150 × 106)
|∆ν | = ∆λ ⇒ ∆λ = |∆ν | = m = 2.00 × 10−13 m
λ 2 c 3 × 108
The resolving power required is then,
λ 632.8 × 10−9
R= = = 3.16 × 106
∆λ 2.00 × 10−13
(c) The mode number is,
2R 2 (3.16 × 106)
m= √ = √ = 1006
π F π 3.996 × 106
The plate spacing is
m λ (1006) (6.328 × 10−5 cm)
d= = = 0.0318 cm
2 2
(d) The free spectral range is,
c 3 × 108 m/s
ν fsr = = = 4.72 × 1011 Hz
2 d 2 (3.18 × 10−4 m)
Expressed as a wavelength range,
λ2 (632.8 × 10−9)2
4.72 × 1011 m = 6.300 × 10−10 m = 0.63 nm

λ fsr = ν fsr =
c 3 × 108
(e) The minimum resolvable wavelength difference is
λ2 λ2 λ2 (632.8 × 10−9)2
∆λm in = = √ = √ = √ = 2.00 × 10−13 m
2 d F 2 d π F /2 dπ F (3.18 × 10−4 m)π 3.996 × 106
8-10. (a) The mode number is,

2 n d (2) (4.5)(2 cm)
m= = = 330, 000
λ 5.46 × 10−5 cm
(b) From Eqs. (8 27) and (8 28),

Tm ax − Tm in Tm ax T 4 r2 4 (0.9)
F= = − 1 ⇒ m ax = 1 + F = 1 + =1+ = 361
Tm in Tmin Tm in (1 − r 2)2 (1 − 0.9)2
(c) The resolving power is
2dF √ √
R= = m F = m π F /2 = (330, 000) π 360 /2 = 9.84 × 106
λ
8-11. The spacing d at which overlap occurs is, d = m λ2/2 = (m + 1) λ1/2. Thus,
m λ2 − λ1) = λ1
2d
∆λ = λ
λ
λ 2
(490 × 10 )
−9 2
d= = m = 0.0218 m = 2.18 cm
2 ∆λ 2 (0.0055 × 10−9)
8-12. The bright bands due to white light satisfy, m λ = 2 d. Thus,

m1 λ1 = m2 λ2 = 2 d and m1 − m2 = 149
These equations solved simultaneously gives m2 = 588.7 → 588. Then,
d = m2 λ2/2 = (588) (546.1 × 10−9 nm)/2 = 0.161 mm
44
8-13. The irradiance as a function of phase has the form I = 4I0 cos2(δ/2).
(a) The fringe maximum occur for,
cos δm/2 = ± 1, δm/2 = m π ⇒ δm = 2 m π
Here m is an integer or zero. Thus the fringe maxima are separated by δm+1 − δm = 2 π.
(b) At the half maxima, δ1/2 = δm ± δFW /2
cos2(δ/2) = 1/2
cos2((δm ± δFW )/2) = cos2((2 m π ± δFW /2)/2) = cos2(δFW /4) = 1/2
√
cos(δFW /4) = ± 1/ 2
|δFW /4| = π/4 ⇒ δFW = π
δm+1 − δm 2 π
(c) The finesse is then, F = = = 2.
δFW π
8-14. Case 1: The irradiances of the emerging beams are
I1 = (0.2) (0.8) I0 = 0.16 I0 and I2 = (0.8) (0.2) I0 = 0.16 I0, I1 = I2
The visibility is
√ √ √
Im ax − Im in I1 + I2 + 2 I1 I2 − I1 + I2 − 2 I1 I2 2 I1 I2 2 I2
V= = √ √ = = =1
Im ax + Im in I1 + I2 + 2 I1 I2 + I1 + I2 − 2 I1 I2 I1 + I2 2 I2
Case 2: In this case,
I1 = (0.2) (0.2) I0 = 0.04 I0 and I2 = (0.8) (0.8) I0 = 0.64 I0, I2 = 16 I1
The visibility is,

p
2 16 I12 8
V= = = 0.47
I1 + 16 I1 17
8-15. (a) Following the method of Section 8 5 and using Eq. (8 22):
r t2e−iδ − r + r (r 2 + t2) e−iδ

+
E0R = − r E0I + r t e−iδE01 = −r+ −iδ
E0I = E0I
1−r e 2 1 − r2e−iδ
r(1 − e−iδ) 2 i r e−iδ/2sin δ/2
E0R = E0I =
1 − r 2e−iδ 1 − r2e−iδ
∗
E0RE0R 4r 2sin2 δ/2
R= ∗ =
E0I E0I (1 − r 2)2 + 4r 2sin2 δ/2
Note that I have assumed lossless mirrors in setting r 2 + t2 = 1.
4r 2sin2 δ/2 + (1 − r 2)2

(b) R + T = =1
(1 − r 2)2 + 4r 2sin2 δ/2
8-16. Using the relation between the finesse and the ratio of the maximum and minimum transmission,
Tmax − Tm in Tm ax T 4 r2 4 (0.6)
F= = − 1 ⇒ m ax = 1 + F = 1 + =1+ = 16
Tm in Tmin Tm in (1 − r 2)2 (1 − 0.6)2
45
+ +√
8-17. In this case, in steady state, Eq. (8 21) should be written as, E01 = t1 E0I + E01 R1 R2 e−iδ
where δ = 2 kd + ϕ1R + ϕ2R and r1,2 = R1,2 e−iϕ1,2R. Rearranging,
p
+ t1
E01 = √ E0I
1− R1 R2 e−iδ
The complex amplitude of the transmitted field is then obtained in a manner that mirrors the development
leading to Eq. (8 23),
t1 t2e−iδ/2
E0T = √ E0I
1 − R1 R2 e−iδ
Then,
I E E∗ |t1|2 |t2|2 T1 T2
T = T = 0T 0T ∗ = √ = √ √
II E0I E0I 1 + R1 R2 − 2 R1 R2 cosδ (1 − R1 R2 )2 + 4 R1 R2 sin2(δ/2)
Here T1,2 = t1,2|2. If the mirrors are lossless, then T1,2 = 1 − R1,2 and R = 1 − T .

FSR
8-18. (a) Estimating from Figure 8 15, F = ≈ 70 (b) ν fsr = c/2d = (3 × 108)/(2 · 0.1) Hz = 1.5 GHz
FWHM
νfsr 1.5 GHz ν 4.53 × 1014
(c) FWHM = = = 21 MHz (d) Q = = = 2.2 × 107
F 70 FWHM 2.1 × 107
(e) τ p ≈ Q/(2 π ν) = (2.2 × 107)/ 2 π · 4.53 × 1014 ) ∼ 8 × 10−9 s
8-19.
The transmittance function has the form, 1
0.9
1
T= 4 F2
0.8
1+ π2
sin2(k d) 0.7
0.6
For the case at hand, 0.5
T
0.4
F = 15 and k = 2 π/λ = 2 π/(632.8 × 10−9 m) = 9.93 × 106/m. 0.3
0.2
0.1
The plot is shown to the right. 0
0 0.2 0.4 0.6 0.8 1
Change in d (µm)
8-21. (a) The peak separation for a given wavelength is d fsr = λ/2 ⇒ λ = 2 ∆d = 2 · 2.95 µm = 5.9 µm
∆d 0.01 × 10−6
(b) Using Eq. (8 37), ∆λ ≈ λ= 5.9 µm = 1.2 × 10−6 µm
d 0.05
2 2 2 m1 d2 − 2 m2 d1 2 m1 (d1 + ∆d) − 2 (m1 + 1) d1 2 m1 ∆d − 2 d1
(c) λ2 − λ1 = d2 − d1 = = ≈
m2 m1 m1 m2 m1 (m1 + 1) m21
Now, m1 = 2 d1/λ ≈ 2 d/λ. So,
∆d λ2 5.92

λ2 − λ1 = λ − = 1.2 × 10−6 µm − µm = − 3.5 × 10−4 µm.
d 2d 2 · 5 × 10−4
8-22. (a) From the plot, 28 divisions ≈ 11.6 GHz. The separation of the absorption dips is then,
11.6 GHz
∆νdips ≈ (25.5 divisions) = 10.6 GHz
28 divisions
(b) Similarly, the full widths at half depth are about 2 divisions and 1.5 divisions leading to widths of
about 0.83 GHz and 0.62 GHz, respectively.
46
8-23. The FWHM of the transmittance peaks is determined as in the text. Picking up the development with Eq.
(8 33),

2π 2π
δ1/2 = − 2d = π/F ⇒
λ λ + δλ1/2
1 1 λ + δλ1/2 − λ 1
− = =
λ λ + δλ1/2 λ (λ + δλ1/2) 4 d F
δλ1/2 λ λ λ
≈ δλ1/2 = =
1 + δλ1/2/λ 4dF 2mF
2 δλ1/2 ≈ λ/(mF )
47
Chapter 9 Coherence
9-1. The given

function is
− 1, − L/2 < x < 0
f (x) = Odd function: a0 = an = 0
+ 1, 0 < x < L/2
Its Fourier series representation is formed as,
∞
2 +L/2
X Z
f (x) = b m sin mkx, where bm = f (x) sin mkxdx =
L −L/2
m=1
(Z )
0 Z L/2
2 2 n 0 L/2
o
bm = − sin mkx dx + sin mkx dx = [cos (m k x)]−L/2 − [cos (m k x)]0
L −L/2 0 mkL

2 mkL mkL 4 mkL 2π 2
bm = 1 − cos − cos +1 = 1 − cos with k = , bm = (1 − cos m π)
mkL 2 2 mkL 2 L mπ
∞
2 X 1 − cos m π

2πmx 2 2 πx 2 6π x
f (x) = sin = 2 sin + sin + ...
π m L π L 3 L
m=1
4 sin (k x) sin (3 k x) sin (5 k x)

or f (x) = + + + ...
π 1 3 5
9-2. The function is,





 E cos ω t,
 3 5
0 π, ω t, π
f (t) = 2 2
5 7


0, <ωt< π


2π 2

The Fourier series representation is formed as:

∞
2 t0 +T
Z Z t0 +T
a0 X 2
f (t) = + am cos mωt where a0 = f (t)dt and am = f (t) cos mωt dt
2 T t0 T t0
m=1
#5π/2
Z t0 +T Z
2 2 2E0 2 E0
a0 = f (t) d t = E0 cos (ω t) d(ω t) = sin ω t =
T t0 ωT ωT π
3π/2
Z 5π/2
2 E0
am = cos ωt cos mωtd(ωt)
ωT 3π/2
Z 5 π/2 5 π/2
2 E0 2 E0 ω t 2 ω t E
m = 1: a1 = cos2ωt d(ω t) = + = 0
ωT 3 π/2 2π 2 4 3π/2 2
5π/2 5π/2
2 E0 sin[(m − 1) ω t] sin[(m + 1) ω t] 2 E0 sin ω t sin 3 ω t

2 E0
m > 1: am = + , m = 2: a2 = + =
ωT 2(m − 1) 2(m + 1) 3π/2 2 π 2 6 3π/2 3π
5π/2
E0 sin 3ωt sin 5ωt

2 E0
m = 3: a3 = a5 = a7 = ... = 0, m = 4: a4 = + =−
π 6 10 3π/2 15 π
All bm = 0 since function is even or, more tediously, by integration.
E0 E0 2 E0 2 E0
Thus f (t) = + cos ωt + cos 2ωt − cos 4ωt + ...
π 2 3π 15 π
48
9-3. The Fourier transform is formed as,

∞ ∞ ∞ t2
− −iωt
Z Z Z
1 1 2/ 2 h 2σ2
g(ω) = f (t)eiωtdt = h e−t 2σ
eiωtdt = e dt
2π −∞ 2π −∞ 2π −∞
Completing the square:

2
t2 σ 2ω 2

t iσω
−iωt= √ − √ +
2 σ2 2σ 2 2
Then,
" 2 #
2
σ 2 ω2

t iσω
∞ − −√ + ∞ t iσω
h σ22ω2 − √ −p
Z √ Z
h 2σ 2 2
2σ 2)
g(ω) = e dt = e e dt
2π ∞ 2π −∞
t iσω dt dt
Let, x = √ − √ so dx = √ , so dx = √ . Then,
2σ 2 2σ 2σ
h −σ22ω2 ∞ −x2√ h −σ22 ω2 √ h −σ22ω2 √
Z ∞
√
Z
2
g(ω) = e e σ dx = e 2σ e−x dx = e 2σ π
2π −∞ 2 π −∞ 2 π
σh −σ 2ω2/2
g(ω) = √ e
2π
9-4. In general,
Z ∞ Z ∞
1
f (t) = g(ω)e iωt
dω where g(ω) = f (t)eiωt dt
−∞ 2π −∞
For the given function,

iωt τ0/2 " #
Z τ0/2 iωτ0/2 −iωτ0/2
1 A e A e − e A ωτ Aτ
0
g(ω) = A eiωt dt = = = sin = 0 sinc (τ0/2)
2π −τ0/2 2 π i ω −τ0/2 π ω 2i πω 2 2π
2 A2τ02
|g(ω)| = sinc2(ω τ0/2)
4 π2
2
|g(ω)|
From the plot,

2π 1
∆ω ≈ ⇒ 2 π ∆ν ≈ 2 π/τ0 ∆ν ≈
τ0 τ0
4π 2π 2π 4π
- -
τ0 τ0 τ0 τ0
∆ω
(590)2 (590)2
9-5. Generally, lt = λ2/∆λ. Then, (lt)1 = = 3.48 × 10−4 m and (lt)2 = = 3.48 × 10−5 m.
100 10
Use longer coherence − length light, thus shorter − bandwidth filter.
9-6. The coherence length is lt = cτ0 = c(10−6 s) = 300 m. The bandwidth is, ∆ν = 1/τ0 = 106 Hz. The resultant
linewidth is ∆λ = λ2/lt = (632.8 × 10−9m)2/(300 m) = 1.33 × 10−15 m.
49
9-7. The angular diameter of the sun viewed from the earth is θ = 0.5◦ = 0.008727 rad. The spatial coherence
width is then,
1.22 λ 1.22(550 × 10−7cm)
ls < = = 0.00769cm
θ 0.008727
If ls is the diameter of a circle of coherence, then “good” coherence exists over an area of 10% the area, or a
√ √
diameter of 0.1 ls < .1 (0.0769 mm) = 0.0243mm.
λ2 (643.8 nm)2 l 30cm

9-8. (a) ∆λ = = = 0.00138nm (b) τc = t = = 1 ns
lt 30 × 107 nm c c
λ2 (5000 Å)2
9-9. The coherence length of the transmitted light is lt = = = 0.25 cm = 2.5mm.
∆λ 1 Å
9-10. The linewidth is for the 0.2 mm exit slit is ∆λ = (20 Å/mm) (0.2 mm) = 4 Å. The coherence length and time
are then,
λ2 (5 × 10−5cm)2
lt = = = 0.0625cm
∆λ 4 × 10−8 cm
τ0 = lt/c = 2.08 × 10−12 s
Lens 2
Lens 1
Prism Slit 2
Slit 1
9-11. See Figure 9 14 in the text for the arrangement. The slit separation a must be within the spatial coherence
width of the light emerging from the pinhole. The distance from the pinhole to the slits is r = 1 m. If the
diameter of the pinhole is s = 5 × 10−4 m, then the angle subtended by the pinhole from the plane con
taining the two slits is θ = s/r. Since the pinhole is circular the spatial coherence width is given by Eq. (9
38). Therefore,
1.22 λ 1.22 r λ 1.22(100 cm)(5.89 × 10−5 cm)
a < ls = = = = 0.144 cm
θ s 0.05 cm
λ2 (6328 × 10−10 m)2 1 c c

9-12. (a) ∆λ = = = 4 × 10−7 Å , ∆ν = = = 4 = 3 × 104 Hz
lt 104 m τ0 lt 10 m
9-13. (a) The linewidth is for the 0.2 mm exit slit is ∆λ = (20 Å/mm) (0.2 mm) = 4 Å. The coherence length and
time are then (as in problem 9 10),
λ2 (5 × 10−5cm)2
lt = = = 0.0625cm
∆λ 4 × 10−8 cm
τ0 = lt/c = 2.08 × 10−12 s
0.4mm
(b) The normalized correlation function is |γ(τ )| = 1 − τ /τ0. But, τ /τ0 = ∆l/lt = = 0.64. So,
0.625mm
V = |γ(τ )| = 1 − τ /τ0 = 0.36.
Imax − Im in 100 − Im in
(c) V = = 0.36 = ⇒ Im in = 47.06.
Imax + Im in 100 + Im in
50
9-14. The coherence length lt and spatial coherence width ls are the length and diameter of the cylindrical coher
ence volume. Now,
λ2 5502
lt = = nm = 1.0083 × 10−4cm = 1.83λave
∆λ 700 − 400
1.22λ 1.22(550)
ls = = nm = 76890 nm = 7.689 × 10−3 cm
θ 0.008727
“Good” ls = 0.25(7.689 × 10−3) = 1.922 × 10−3cm = 35 λave
The area of the base of the cylindrical coherence volume is
π ls2 π
Abase == (1.922 × 10−3)2 = 2.90 × 10−6 cm2
4 4
√
9-15. (a) For unequal beams and partial coherence, I = I1 + I2 + 2 I1 I2 Re(γ) where,
Re(γ) = Re (1 − τ /τ0)eiωτ = (1 − τ /τ0) cos(ω τ ) = |γ(τ )| cos(ω τ )

So,
√
Im ax = I1 + I2 + 2 I1 I2 |γ(τ )|
√
Im in = I1 + I2 − 2 I1 I2 |γ(τ )|
√ √ √
Im ax − Imin I1 + I2 + 2 I1 I2 |γ(τ )| − I1 + I2 − 2 I1 I2 |γ(τ )| ) 2 I1 I2 |γ(τ )|
V= = √ √ =
Im ax + Imin I1 + I2 + 2 I1 I2 |γ(τ )| + I1 + I2 − 2 I1 I2 |γ(τ )| ) I1 + I2
(b) For beams of the same irradiance √
2I1 I1 |γ(τ )|
Veq = = |γ(τ )|
I1 + I2
When V = 0.9 Veq = 0.9|γ(τ )| , we have,
√
2 I1 I2 |γ(τ )|
= 0.9|γ(τ )| ⇒ 4 I1 I2 = 0.81 (I1 + I2) with R = I1/I2,
I1 + I2
0.81 R2 − 2.38 R + 0.81 = 0 ⇒ R = 2.55
τ ∆ ∆ mλ m ∆λ
9-16. The optical path difference in mth order is ∆ = m λ. Then, = = = =
τ0 lt λ2/∆λ λ2/∆λ λ
m ∆λ
As shown in problem (9 15b) for equal irradiance beams, V = |γ(τ )| = 1 − τ /τ0 = 1 −
λ
9-17. (a) lt = λ2/∆λ = (546.1)2/0.05 = 5.965 × 106 nm

m ∆λ 20 (0.05)
(b) With a filtered Hg lamp, V = 1 − =1− = 0.998.
λ 546.1
20 (10)
With filtered white light, λave = 546 nm and ∆λ = 10nm, V = 1 − = 0.63
546
∆λ (∆) (∆λ) (2 × 107) (.0013)

9-18. ∆ = m λ = 2 × 1 cm = 2 × 107 nm, V = 1 − m =1− 2
=1− = 0.937
λ λ (643.847)2
(108) (0.0013)
For a 5 cm movement, ∆ = 2 × 5 cm = 108 nm, and V = 1 − = 0.686
(643.847)2
(2 d) (∆λ)
V → 0 when 1 − → 0 or when d = λ2/2 ∆λ, so d = 0.0895cm.
λ2
(2 × 107) (0.025)
9-19. (a) d = 1cm: V = 1 − < 0 or V = 0 (See Prob. 9 16). d = 5 cm, V = 0
(546.1)2
λ2 (546.1)2
V = 0 first when d = = = 0.596cm
2 ∆λ 2 (0.025)
2
∆λ λ (0.15) (546.1)2
(b) V = 0.85 = 1 − 2 d 2 ⇒ d = 0.15 = = 0.0895 cm
λ 2 ∆λ 2 (0.025)
51
Chapter 10 Fiber Optics
10-1. The information rate is,

(4000 Hz)(2 × sampling rate) (8 bits/sample) = 64kbps
44.7 Mbps
Number of channels = − 26(assigned) = 672
64 kbps
10-2. Each TV channel requires a bandwidth of 6 MHz.
ν c/λ (3 × 108/1.55 × 10−6)

(a) # of TV stations = = = = 32 million
6 × 10 Hz 6 × 10 Hz
6 6 6 × 106
4 × 1012
(b) #of TV stations = = 0.67 million
6 × 106 Hz
r
n1 2 p
10-3. (a) In general, Ls = d − 1 . When θ = θm , sin θ = sin θm = n21 − n22 . Then,
sin θ
s 2
n1 n d
Ls = −1= p 2
n21 − n22 n21 − n22
(b) In one meter of fiber, there are 1/Ls reflections, so the number of reflections per meter Ns is,
p
n21 − n22 (1.462 − 1.4572)1/2
Ns = = = 1284/m
n2 d 1.457 (50 × 10−6 m)
10-4. (a) 1.52 sin ϕc = 1.41 ⇒ ϕc = 68.1◦ (b) N.A. = n0 sin θm = n1 cos ϕc = 1.52 cos(68.1◦) = 0.567
(c) sin θm = (N.A.)/n0 = (0.567/1) = 0.567 ⇒ θm = 34.5◦
p √
10-5. (a) N.A. = n21 − n22 = 1.532 − 1.392 = 0.64
(b) Since n0 sin θm = N.A., θm = sin−1(0.6394) = 39.7◦. The cone angle is 2 θm = 79.5◦.
(c) The number of reflections per inch is Ns = 1/Ls, so,
1 1 1
Ns = = r = = 184/inch
Ls 2 r 2
n1 1.53
d n sin θ
0
− 1 0.0025 in) sin(39.75 ◦)
− 1
That is, there will be 6624 reflections in 3 ft. At θ = θm/2 = 19.87◦, 1/Ls = 91.14/inch so that there would
be 3281 reflections in 3 ft.
10-6. Refer to Figure 10 3b in the body of the text. In that figure, xs is the hypotenuse of the triangle formed by
θ ′, Ls, and d. Combining a relation that follows from examination of that triangle with Snell’s law gives,
d n1 d n d
xs = = = 1
sin θ ′ n0 sin θ sin θ
(b) The total distance is xt = xs Ns where Ns is the total number of skips. That is,
" #
L n1d L n1 L
xt = xs = =
Ls sin θ d (n1/sin θ)2 − 1 1/2
p
n1 − sin2 θ
2
(c) For the given parameters,

n1 d (1.5) (50 µm)
xs = = = 432 µm
sin θ sin(10◦)
q p
Ls = d n1/sin θ)2 − 1 = (50 µm) (1.5/sin(10◦))2 − 1 = 429 µm
n1 L 1.5 (10 m)
xt = 1/2 =p = 10.07 m
n1 − sin2 θ
2 1.52 − sin2(10◦)
52
10-7. The number of modes is,
2 2
1 πd 2 1 πd 1
n21 − n22 = (π 40/0.85)2 1.4612 − 1.4562 = 159

mm ax = ((N.A.) =
2 λ 2 λ 2
10-8. The necessary condition is,

d 2.4 2.4 λ 2.4 (1.25 µm)
< ⇒d= = √ = 10.2 µm
λ π (N.A.) π (N.A.) π 1.462 − 1.4572
p
10-9. In general the number of √modes is given by N = 1 + (d/λ) n21 − n22 . Then,
For d = 5 λ: N = 1 + 2 (5) 3.62 − 3.552 = 6 (or 12 for two polarizations).
√
For d = 50λ: N = 1 + 2 (5) 3.62 − 3.552 = 60 (or 120 for two polarizations).
10-10. The loss in db/(100 m) is

αdb = 10 db/(100 m) log10 (P1/P2) = (0.1 db/m) log10 (5) = 0.070 db/m = 70 db/km
10-11. The entire cable has a loss of 3 × 5 db + 2 db = 17 db in 3 km. So αdb = 17db/3km. Then, using Eq. (10 13):

αdb = (10 db/km) log10 (P1/P2) ⇒ P2 = P1 10−αd b L/10(db/km )

P2 = (4 mW 10−[(17/3)3/10] = (4 mW 10−17/10 = 0.080 mW

10-12. The loss in db that can be tolerated in a length L of fiber is,

αdb L = (10 db) log10 (0.01/10−6) = 40 db
So for the coaxial cable: L = (40 db/12 db/km) = 3.33 km. For the fiber: L = (40 db/4 db/km) = 10 km.
10-13. Rayleigh scattering has a 1/λ4 dependence so,

α1.55 = α0.90 0.9/1.55)4 = (1.2 db/km) 0.9/1.55)4 = 0.136 db/km
10-14. (a) The fractional power loss is

P1 − P2
f= = 1 − P2/P1 ⇒ P2/P1 = 1 − f
P1
Then,
1
αdb = (10 db/km) log10 = − (10 db/km) log10 (1 − f )
1−f
(b) Using the relation from part (a),For f = 0.25, αdb = 1.25 db/km, for f = 0.75, αdb = 6.02 db/km ,
forf = 0.90, αdb = 10db/km and for f = 0.99, αdb = 20 db/km
10-15. (a) The ratio of the longest and shortest path lengths is,
Lshort n2 1.45
= =
Llong n1 1.46
so,
Llong = Lshort (1.46/1.45) = (1 km)(1.46/1.45) = 1.0069 km
(b) The transit times are ,
Lshort n1 1.46
tshort = = Lshort = (1 km) = 4.867 × 10−6 s
v c 3 × 105 km/s
L n 1.46
tshort = short = 1 Lshort = (1.0069 km) = 4.900 × 10−6 s
v c 3 × 105 km/s
53
10-16. First find the index of refraction of the cladding:
p p p
sin θm = n21 − n22 ⇒ n2 = n21 − sin2 θm = 1.4462 − sin2(35◦) = 1.3274
Then,
τ n n −n 1.446 1.446 − 1.3274

δ = 1 1 2
= = 431 ns/km
L c n2 3 × 10−4 km/ns 1.3274
For non overlapping pulses,
δ(τ ) > T = 1/ν ⇒ νm ax = 1/δ(τ ) = 1/(431 ns) = 2.32 MHz
10-17. The entrance angle θ = 15◦ is not necessarily θm so Eq. (10 14),
! " #
∆L n1 n1 n1 n1 (1.48) (1 km) 1.48
δ(τ ) = = ∆L = (xt − L) = L p −1 = −1
v c c c 3 × 10−4 km/ns
p
n21 − sin2θ 1.482 − sin2(15◦)
δ(τ ) = 77.2 ns
Here the expression for xt is given by problem 10 6b.
10-18. The group delay is

1.46

n n1 − n2
δ(τ /L) = 1 = (0.003) = 14.6 ns/km
c n2 3 × 10−4 km/ns
10-19. Using Eq. (10 15) the refractive index profiles is seen to be,
n(r) = n1 1 − 2 ∆(r/a)α = 1.5 1 − (0.02) (r/50 µm)α
p p
1.502
1.5
1.498
1.496
1.494
n(r)
1.492
1.49
1.488
1.486 α=2
α=10
1.484
0 10 20 30 40 50
r (µm)
10-20. The relative index difference is,

n1 − n2 1.46 − 1.44
∆= = = 0.0137
n2 1.46
Using Eq. (10 16) and the relation that follows,
n1 L 2 1.46 (1 km)
δ(τ )G = ∆ = (0.0137)2 = 457 ps
2c 2 (3 × 10−4 km/ns)
2 2
δ(τ )SI = δ(τ )G = δ(τ )G = 146 × 457 ps = 66.7 ns
∆ 0.0137
0.5 0.5 0.5

10-21. νmax L = ⇒ νm ax = = = 25 MHz
δ(τ /L) L δ(τ /L) (1 km) (20 ns/km)
10-22. At λ = 820 nm, Figure 10 13 gives M ≈ 100 ps/nm km. Then δ(τ /L) = − M ∆λ. So,
For the LED: δ(τ ) = − M ∆λ L = (100) (40) (1) ps = 4000 ps = 4 ns.

For the LD: δ(τ ) = − M ∆λ L = (100) (4) (1) ps = 400 ps = 0.4 ns.
54
10-23. The total time delay is found as
p
δτ = δτm2 o d + δτm2 at
where,
L n1 (1 km) (1.46)
δτm o d = ∆= (0.01) = 48.7 ns
c 3 × 10−4 km/ns
δτm at = M L ∆λ = (100) (1) (40) ps = 4 ns
So,
√
δτ = 48.72 + 42 ns = 48.9 ns
10-24. (a)The plot is shown below,
8
Wavegu de d spers on (ps/km/ns)
7 (b) From the plot:

M ′ ≈ 3.9 ps/km nm at 1.27µm
6
M ′ ≈ 4.3 ps/km nm at 1.55µm
5 Then,
At 1.27µm:
4 ps ps
∆(τ /L) = M ′ ∆λ = (3.9) (1) = 3.9
km km
3 At 1.55µm:
ps ps
∆(τ /L) = M ′ ∆λ = (4.3)(1) = 4.3
2 km km
1
0.6 0.8 1 1.2 1.4 1.6 1.8
Wavelength in micrometers
10-25. (a) Modal:

(1.47)(1) 1.47 − 1.455

n1 L n1 − n2
δτm o d = = ns = 50.5 ns
c n2 3 × 10−4 1.455
Material:
δτm at = L M ∆λ = (1) (43) (25) ps = 1.075ns
Waveguide:
δτWG = L M ′ ∆λ = (1) (3) (25) ps = 0.075 ns
(b) The total is then,
p
δτ = δτm2 o d + δτm2 at + δτWG
2
≈ 50.5ns
10-26. Consider Figure 10 16a in the body of the text. Let the phase difference due to the differing lengths of the
paths of the two beams be k ∆L. Then the phase difference for the light exiting through Output 1 is
∆ϕ1 = k ∆L and that for the light√ exiting through
√ Output 2 is ∆ϕ2 = k ∆L + π, where the addition of π is
due to the fact that r1′ = − 1/ 2 = eiπ 1/ 2 = eiπ r2. Now the transmission coefficients at each beam
√
splitter must be t1 = t2 = 1/ 2 so that the amplitudes each of the four individual beams exiting through an
output port is
p p 1
E0,out = (1/ 2) (1/ 2) E0,in = Ein
2
The irradiance of each of the four individual output beams is then, Iout = Iin /4. The total irradiance at
each output port is then given by the two beam, equal irradiance, interference relation,
I I
IT , port 1 = 2 Iout (1 + cos(∆ϕ1)) = 2 in [1 + cos(k ∆L)] = in [1 + cos(k ∆L)]
4 2
Iin I
IT , p ort 2 = 2 Iout (1 + cos(∆ϕ2)) = 2 [1 + cos(k ∆L + π)] = in [1 − cos(k ∆L)]
4 2
Thus the total irradiance through the two outputs is, IT = IT , port 1 + IT , p ort 2 = Iin .
55
∆λ 0.8 3 × 108
10-27. (a) The frequency difference is ∆ν = ν= Hz = 10 0.0 GHz
λ 1550 1550 × 10−9
(b) 40 × 10 GHz = 4 THz.
10-28. (a) The conditions to be satisfied are,
∆L = m λ1n = (m + 1/2) λ2n
Here λ1,2n = λ1,2/n = λ1,2/1.5. No ∆L satisfies this relation with exactly integer m but a reasonably good
solution is
∆L = 950.01 λ1n = (950.01) (1550.8 nm)/1.5 = 950.500 λ2n
Many other approximate solutions exist.

(b) Since,
∆L 950.01 λ1n 1550.8

= = 950.01 = 949.5201778808971
λ3n λ3n 1551.6
is close to an integer +1/2. This light will exit primarily through Output 2.
(c) The total irradiances are given by the expressions in the solution to problem 10 26:
IT ,2 1 − cos(2 π∆L/λn)
R= =
IT ,1 1 + cos(2 π ∆L/λn)
Here,
2 π ∆L/λn = 2 π(950.01) λ1n/λ2n = (2 π) (950.01) (1550.8/1550.4) = 1900.51 π
and
R = 1.065 ⇒ 1/R = 0.94
56
Chapter 11 Fraunhofer Diffraction
11-1. See Figure 11 18 that accompanies the problem in the text for a sketch of the setup. The minima are
located as position, ym determined as,
m λ = b sin θm = b ym/f ⇒ ym = m λ f /b
(a) The first minimum occurs at y1 = λ f /b = (546.1 × 10−6 mm) (60 cm)/(0.015 cm) = 2.18 mm.
(b) The separation of the first and second minimum is
y2 − y1 = (2 − 1)λ f /b = 2.18 mm
11-2. Refer to Figure 11 1 and the surrounding discussion. The ratio I/I0 is given be Eq. 11 10 is
sin2 β
I/I0 =
β2
1
where β = k b sin θ. The path difference from the edges of the slit to a point on the screen is ∆ = b sin θ.
2
So when ∆ = 3/4 λ,
1 1 2π 3
β= k (3/4) λ = λ = 3 π/4
2 2 λ 4
Under this condition,
I sin2(3 π/4)
= = 0.090
I0 (3 π/4)2
11-3. See Figure 11 19 that accompanies the problem in the text.

(a) The diffraction minima are located at angles θm = ym/L where L = 2 m is the slit to screen distance,
The positions of the minima are given by m λ = b sin θm = bym/L ⇒ ym = mλL/b. Then,
6 λ L 6 (632.8 × 10−7 cm) (200 cm)

y3 − y−3 = ∆y = (3 − ( − 3))λ L/b ⇒ b = = = 0.013 cm = 0.13 mm
∆y 5.625 cm
(b) Lm in = b2/2λ, so,
L 200 cm
= = 139
Lm in (0.0135 cm)2/(2 · 632.8 × 10−7 cm)
The screen is in the far field.
11-4. Let m1 = 5 for λ1 and m2 = 4 for λ2. Then,
m1 λ1 = m2 λ2 = b sin θ
5 λ1 = 4 λ2 = 4 (620 nm) ⇒ λ1 = 496 nm
11-5. Let the full angle breadth between the first minim on either side of the central maximum be ϕ = 2 θ, where
θ is the angle that locates the first minimum relative to the center of the pattern. For m = 1,
λ 550 nm
λ = b sin θ = b sin (ϕ/2) ⇒ b = =
sin(ϕ/2) sin(ϕ/2)
For ϕ = 30◦, b = 2.125 µm, for ϕ = 45◦, b = 1.437 µm, for ϕ = 90◦, b = 0.778 µm, for ϕ = 180◦, b = 0.55µm.
57
(1) (550)
11-6. (a) sin θ = m λ/b = ⇒ θ = 15◦
2125
kb πb π (2.125 µm)
(b) β = sin θ = sin θ = sin θ = 12.138 sin θ. Then
2 λ 0.55 µm
I/I0 = sin2 β/β 2
θ=5 : ◦
β = 1.058 rad, I/I0 = 0.678
θ = 10◦: β = 2.108 rad, I/I0 = 0.166
θ = 15◦: β = 3.142 rad, I/I0 ≈ 0
θ = 22.5◦: β = 4.645 rad, I/I0 ≈ 0.0461
11-7. (a) The fourth and fifth intersections of β = tanβ occur at 4.477 π and 5.482 π.
Iβ =4.477 π sin2(4.477 π)/(4.477 π)2 1 I β=5.482 π sin2(5.482 π)/(5.482π)2 1

(b) = = ; = =
Iβ =0 1 199 Iβ =0 1 298
11-8. For the single slit:

I1 sin2(1.43 π)/(1.43π)2
= = 0.047
I0 1
I2 sin2(2.46 π)/(2.46π)2
= = 0.017
I0 1
For the circular aperture: (See Figure 11 8b)
I1
= 0.0175
I0
I2
= 0.00416
I0
(1706.9) (5.5 × 10−5)

k πD πD y fλγ
11-9. γ = D sin θ = sin θ = ⇒y= =γ cm = (3.268 × 10−4 cm) γ
2 λ λ f πD 91.44π
1st secondary maximum (γ = 5.15): y = 1.68 × 10−3 cm
2nd secondary maximum (γ = 8.42): y = 2.75 10−3 cm
11-10. 1.22 D sin θ = D y/f ⇒ y = R = 1.22 λ f /D = (1.22)(5.5 × 10−5 cm)(150)/12 = 8.39 × 10−4 cm
11-11. Using Eq. (11 21) the angular half width of the Airy disc formed on the moon will be,
1.22 λ
∆θ1/2 =
D
where D is the diameter of the circular aperture. The radius R of the airy disc formed on the moon, which
is a distance L from the aperture is
1.22 λ L 1.22 (10.6 × 10−6 m)(3.76 × 108 m)
R = L tan ∆θ1/2 ≈ L ∆θ1/2 = = = 4.86 × 106 m
D 10−3 m
The diameter of the laser spot on the moon is about 9.72 × 106 m. The irradiance in the spot (assuming a
nearly constant irradiance over the spot (this is not really the best approximation, but it gives an order of
magnitude estimate),
Φ Φ 2000 W
I= = = = 2.7 × 10−11 W/m2
A π R2 π (4.86 × 106 m)2
58
11-12.
The angular radius θ of the first minimum in the diffraction pattern is
given by the relation, sin θ ≈ θ = 1.22 λ/D. Then from the geometry
θ illustrated in the figure,
D 2D
D D2 (2 × 10−3 m)2
θ≈ = 1.22 λ/D ⇒ L = = = 5.18 m
L 1.22 λ 1.22 (632.8 × 10−9 m)
L
11-13. The distance L for the headlights to be barely resolvable if they are separated by a distance y is given be
Eq. (11 22), as,
yD (45 × 2.54 cm)(0.5 cm)
∆θmin = y/L = 1.22 λ/D ⇒ L = = = 8.517 × 105 cm = 27, 900 ft = 5.3 miles
1.22 λ 1.22 (5.5 × 10−5 cm)
11-14. (a) The minimum and maximum distances are for a line separation of s = 1mm is
s Dmin (10−3) 2 × 10−3

s 1.22 λ
Min: ∆θ = = ⇒L= = m = 3.0 m
L D 1.22 λ 1.22 · 550 × 10−9
s Dmax (10−3) 7 × 10−3

Max: L= = m = 10.4 m
1.22 λ 1.22 · 550 × 10−9
(b) The pupil diameter will be
1.22 λ L 1.22(550 × 10−9)L
D= = = 6.71 × 10−4 L
s 10−3
11-15. (a) According to Eq. (11 30) the condition for missing orders is, a = (p/m) b. The fourth order interference
maxima are missing so p = 4 m and a = 4 b = 4 (0.1 mm) = 0.4 mm.
(b) The irradiance is given by,
2
sin β

I = 4 I0 cos2 α
β
In zeroth order I = 4 I0. The interference maxima occur for,

p λ = a sin θ ⇒ sin θ = 0, ± λ/a, ± 2 λ/a, ± 3 λ/a, ...
Also cos2 α = 1, so at the interference maxima
2
sin β

I = 4 I0
β
where β = (k b/2) sin θ = (π b/λ) sin θ.Then:
2 2
sin β sin (π/4)

πb λ I
p = 1: sin θ = λ/a; β = = π (b/a); = = = 0.8106
λ a 4 I0 β π/4
2 2
sin β sin (π/2)

πb 2λ I
p = 2: sin θ = 2λ/a; β = = 2π (b/a); = = = 0.4053
λ a 4 I0 β π/2
2 2
sin β sin (3 π/4)

πb 3λ I
p = 3: sin θ = 3λ/a; β = = 3π (b/a); = = = 0.0.0901
λ a 4 I0 β 3 π/4
11-16. (a) If a/b = N where N is an integer, the the N th fringe is missing on either side of the central maximum.
If a/b falls between N − 1 and N , then the (N − 1)th fringes appear on either side. The total number of
fringes is (including the central maximum) is thus 2 (N − 1) + 1 = 2 (a/b − 1) + 1 = 2 (a/b) − 1.
(b) For 13 bright fringes in the central diffraction peak,

2 (a/b) − 1 = 13 ⇒ a/b = 7 ⇒ a = 7 b = 7 (0.3 mm) = 2.1 mm
59
11-17. (a) Diffraction minima occur for,
m λ = b sin θ ⇒ m λ ≈ b θ
Thus the angular width of the central (m = 1) diffraction peak is (∆θ1/2)diff = λ/b.
Interference minima occur for,
(p + 1/2) λ = a sin θ ≈ a θ ⇒ (∆θ1/2)int = λ/2 a (p = 0)
Thus,
(∆θ1/2)diff λ/b
= = 2 (a/b)
(∆θ1/2)int λ/2a
(b) For a = 10 b:
peak width
= 2 (a/b) = 20 ×
fringe width
11-18.
Consider the geometry shown at left. Begin with an expression like Eq. (11 4)
s but for three slits. Then,
(Z b Z −a+ b Z a+ b )
EL i(kr0 −ωt) 2
ikssin θ
2
ikssin θ
2
ikssin θ
b EP = e e ds + e ds + e ds
r0 −b
−a−
b
a−
b
2 2 2
a
Taking the integrals in turn:
b s=0 Z b
2 k
a eikssin θds = b sin(β)/β as for the single slit in the text. Here β = b sin θ.
−b 2
2
b
−a+ 2
e2iα
Z
1 h iksin θ(a+b/2) i
eikssin θds = e − eiksin θ(a−b/2) = (2 sin β)
−a− 2
b i k sin θ k sin θ
k
where α = a sin θ. The final integral follows from the one just done by letting a → − a, α → − α. That is
2
Z a+ b
2 e−2iα
eikssin θds = (2 i sin β)
a− 2
b i k sin θ
Altogether, then,
e2iα e−2iα

EL i(kr0 −ωt)
EP = e b sin(β)/β + (2 sin β) + (2 sin β)
r0 k sin θ k sin θ
b sin(β) 4 sin β cos 2 α b sin(β) 4 sin β cos 2 α

EL i(kr0 −ωt) E
EP = e + = L ei(kr0 −ωt) +
r0 β k sin θ r0 β 2 β/b
ELb i(kr0 −ωt) sin β
EP = e (1 + 2 cos 2 α)
r0 β
2
sin β

2
I = I0 (1 + cos 2 α)
β
This result can be put into the general form by noting,
3 sin α − 4 sin3 α sin 3α
1 + 2 cos 2 α = 1 + 2 1 − 2 sin2 α = 3 − 4 sin2 α =

=
sin α sin α
So,
2 2
sin β sin 3α

I = I0
β sin α
A plot is shown on the next page.
60
11-18. continued . For a = 3 b, and k b = 50, the plot is shown below.
9
5
I/I0
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
sinθ
11-19. For N = 7, α/β = 4

50
45
40
35
30
I/I0
25
20
15
10
0
0 π/4 π/2 3π/4 π
β=α/4
11-20. The irradiance is given by,

2 2
sin β sin N α

I = I0
β α
π
with N = 10, a = 5 b, b = 10−4 cm, λ = 435.8 nm. Recall that β = b sin θ, so α = 4 β. For interference
λ
maxima,
2 2
sin N α sin β

= 1 ⇒ I = I0
α β

πb mλ
Also, sin θ = m λ/a and β = = m π (b/a) = m π/5. Then,
λ a
2 2
sin β sin(π/5)

For m = 1: I/I0 = = = 0.875
β π/5
2 2
sin(2 π/5) sin(3 π/5)

For m = 2: I/I0 = = 0.573. For m = 3: I/I0 = = 0.255
2 π/5 3 π/5
2 2
sin(4 π/5) sin(5 π/5)

For m = 4: I/I0 = = 0.0547. For m = 5: I/I0 = = 0.
4 π/5 5π/5
61
11-21. Consider an N slit aperture, with N an odd number, arranged like the 3 slit aperture of problem 11 18
with the origin at the middle of the central slit. Then, the electric field at point P on the screen has the
form,
(NX
−1)/2 Z − ja+ b
(Z b " Z ja+ b #)
EL i(kr0 −ωt) 2
ikssin θ
2
ikssin θ
2
ikssin θ
EP = e e ds + e ds + e ds
r0 −b
− ja−
b
ja−
b
2 j=1 2 2
The first integral is just the single slit integral simplified in the text. That is,
Z b
2 b sin β k
eikssin θds = , β = b sin θ
−b β 2
2
The sum of the integrals in the square brackets is,

Z −ja+ b Z ja+ b
2 2 e−2ijα(eiβ − e−iβ) + e−2ijα(eiβ − e−iβ ) 4 i (sin β) (cos 2 j α)
eikssin θds + eikssin θds = =
b
−ja− 2
b
ja− 2 i k sin θ i k sin θ
Z − ja+ b Z ja+ b
2 2 2 b (sin β) (cos 2 j α)
eikssin θds + eikssin θds =
− ja−
b
ja−
b β
2 2
k
where α = a sin θ. Then,
2
(NX
−1)/2 b b (NX
−1)/2
"Z #
−ja+ 2 Z ja+ 2
2 b (sin β)
e ikssin θ
ds + e ikssin θ
ds = cos 2jα
−ja− 2
b
ja− 2
b β
j=1 j =1
The sum can be performed as,
(NX
−1)/2 (NX
−1)/2
( (N −1)/2
! ) ( )
2iα (N −1)/2
X (e ) − 1
cos 2jα = Re e2ijα = Re (e2iα) j = Re
e2iα − 1
j =1 j =1 j =1
(NX−1)/2
− e−iα cos Nα + i sin Nα − cos α + i sin α sin Nα 1
iNα
e
cos 2jα = Re = Re = −
eiα − e−iα 2 i sin α 2 sin α 2
j =1
(NX
−1)/2 Z − ja+ b Z ja+ b (N −1)/2
" #
b (sin β) sin Nα

2 2 2 b (sin β) X
e ikssin θ
ds + e ikssin θ
ds = cos 2jα = −1
− ja−
b
ja−
b β β sin α
j=1 2 2 j=1
sin β sin Nα sin β sin Nα

E b E b
EP = L ei(kr0 −ωt) 1− + 1 = L ei(kr0 −ωt)
r0 β sin α r0 β sin α
2 2
sin β sin Nα

I = I0
β sin α
11-22. (a) See Figure 11 21 that accompanies the problem in the text. Proceeding apace,
2 2
sin α sin β

I = I0
α β
k πa y π (0.2) y
α = a sin θ = = = (1.151/mm) y
2 λ f (546 × 10−6) (1000)mm
π (0.1) x
β= mm−1 = (0.575/mm) x
(546 × 10−6) (1000)
2 2
sin (1.151y/mm) sin (0.575 x/mm) sin2 (1.151y/mm)sin2 (0.575 x/mm)

I = I0 = I0
1.151y/mm 0.575 x/mm (0.438 x2 y2)/mm2
(b) Minima occur first at m = 1 in y = m λ f /a = 2.73 mm and n = 1 in x = n λ f /b = 5.46 mm.
(c) As x → 0, (sin β)/β → 1 and as y → 0, (sin α)/α → 1. So,

2 2
sin (1.151) sin (0.575)

At x = 0, y = 1 mm, I = I0 = 0.629 I0 At x = 1 mm, y = 0 , I = I0 = 0.0.895I0
1.151 0.575
2 2
sin (3 × 1.151) sin (2 × 0.575 )

(d) At x = 2 mm, y = 3 mm, I0 = 0.005 I0
3 × 1.151 2 × 0.575
62
11-23. Since m λ = a sin θ, with m = 1, sin θ1/2 = λ/a.
(a) sin θ1/2 = λ/λ = 1 ⇒ θ1/2 = 90◦ (b) sin θ1/2 = λ/5λ = 0.2 ⇒ θ1/2 = 11.5◦
(c) sin θ1/2 = λ/10λ = 0.1 ⇒ θ1/2 = 5.7◦
11-24. Using the large γ expression,

2
I0 sin γ − cos γ

J1(γ)
I = I0 ≈ 2 √
γ γ πγ
Minima occur at I = 0 or sin γ = cos γ. That is for γ = π/4, 5 π/4, 9 π/4... = (4 m − 3) π/4, m = 1, 2, 3, ...
So minima occur for
k πD
γ= D sin θ = sin θ = (4 m − 3) π/4 ⇒ sin θ = (4 m − 3) λ/4 D
2 λ
The angular separation between consecutive orders (∆m = 1) is
4 ∆m λ ∆m λ λ
(cos θ) ∆θ = = ⇒ ∆θ =
4D D D cos θ
11-25. We have I = I0(sin2 β)/β 2. At secondary maxima, β1 = 1.43 π, β2 = 2.46 π, β3 = 3.47 π or approximately,
β1 = 1.5 π, β2 = 2.5 π, β3 = 3.5 π, ... ≈ (m + 1/2) π. So that sin β ≈ ± 1.
2
( ± 1) I0
(a) ThenI = I0 =
β [(m + 1/2)π]2
(b)
Approximate: More Accurate
2
sin(1.43 π)

I0
I1 ≈ = 0.0450 I0 I1 = I0 = 0.0472 I0
(1.5 π)2 1.43 π
2
sin(2.46 π)

I0
I2 ≈ = 0.0162 I0 I2 = I0 = 0.0165I0
(2.5 π)2 2.46 π
2
sin(3.47 π)

I0
I3 ≈ = 0.00827 I0 I3 = I0 = 0.00834I0
(3.5 π)2 3.47 π
11-26. The principal maxima satisfy m λ = a sin θ, so that,
a 2/3 2
mm ax = = =
λ 1 3
Thus m = 1 is not allowed and only the central maximum (m = 0) appears. The beam width is confined by
the first minimum, given by m λ = a sin θ, with,
p ±1 1
m= = =±
N N 3
Then
λ3 1
sin θ = ± = ± ⇒ θ = 30◦
3 2 2
The angular half width of the beam is 30◦.
63
11-27. For three equal slits, the amplitude phasors must be equal and form a closed polygon an equilateral tri
angle. Thus the phase difference must be ϕ = 2 π/3 = 120◦. Or, using Eq. (16 33), I = 0 when α = pπ/N
with p = ± 1, ± 2, ...
π 2π 2π 4π
Then α = ± , ± , ... and since ϕ = 2 α, ϕ = ± ,± , ... = ± 120◦, ± 240◦...
3 3 3 3
The additional values of ϕ also produce equilateral triangles (notice that for b ≈ 0, β ≈ 0 and (sin β)/β ≈ 1.)
(b) If ϕ = π, the phasors alternately add and subtract leaving only one contributing. Thus, at P , I p ∝ E02.
While at the center the phasors all add and so Im ax ∝ (3 E0)2 = 9 E02. So Ip = Im ax /9. Or since ϕ = 2 α ⇒ α =
π/2, IP = I0[(sin (3 π/2)/(π/2)]2 = I0. Compared with Im ax = 9 I0, IP = Im ax /9.
pπ
c) At a principal maximum p = 0, ± N , ± 2N , ... and so α = = 0, ± π, ± 2 π, ...
N
Thus, sin(Nα)/sin(α) = ± N . Then IP = N 2 I0 = Im ax and IP /Imax = 1.
(d) Note that Iav = I on the screen when it is uniformly illuminated with the energy from the three slits.
Thus Iave = 3 I0, as for the case of incoherent light. On the other hand, IP = 9 I0 in the case of coherence.
Thus IP = 3 Iave .
11-28. Principal maxima and zeros of a 4 slit aperture in terms of phasor diagrams:
Principal maxima: α = 0, π, 2 π, ... ⇒ ϕ = 2 α = 0, 2 π, 4 π, ...
E0 2π
Zeros: α = p π/N = ± π/4, ± 2 π/4, ± 3 π/4, ...,p = ± 1, ± 2, ...p but not equal to ± 4, ± 8, ..., ϕ = 2 α = ± π/2,
± π, ± 3 π/2, ...
270◦
90◦
ϕ = 90◦ ϕ = 180◦ ϕ = 270◦
64
Chapter 12 The Diffraction Grating
12-1. The angular positions of constructive interference are described by m λ = a sin θ. For the case at hand m =
2, λ1 = 400 nm, and λ2 = 600 nm. Then,
2 (400 × 10−7)cm

m λ1
θ1 = sin−1 = sin−1 = 23.58◦
a 1/5000cm
−1 2 × 600 cm
−7

−1 m λ2
θ2 = sin = sin = 36.87◦
a 1/5000cm
The angular separation is then,
∆ θ = θ2 − θ1 = 13.292◦ = 13◦18
′
12-2. (a) See Figure 12 13 that accompanies the statement of the problem in the text. With m λ = a sin θ, the
angular dispersion is found as,
dθ m m 3
D≡ = = = 1 = 14366 rad/cm
d λ a cos θ a cos (sin−1 m λ/a)

3 × 650 × 10 −7
cos sin−1
3500 1/3500
rad 180 deg cm
D = 14366 × × 10−7 = 0.0823 deg/nm
cm π rad nm
The linear dispersion is then,
−7 rad

dy dθ
=f = f D = 1500 mm 14366 × 10 = 2.1549 mm/nm
dλ dλ nm
So the plate factor is,

dλ
= 0.464 nm/mm
dy
(b) According to Eq. (12 11) he resolving power is, R = m N = (3) (3500/cm) (6 cm) = 63, 000
12-3. (a) Principal maxima: m λ = a sin θ. So, for the maximum
−1 2 × 600 × 10
−7

−1 m λ
θ1 = sin = sin = 6.8921◦
a 10−3
1
Minima: m ± λ = a sin θ. So for the neighboring minimum,
N
1 1
2 + 24 600 × 10−7

m± N λ
θ2 = sin−1 = sin−1 = 7.0364◦
a 10−3

1
Then, ∆ θ = θ2 − θ1 = 0.1443◦ = 8.7 ′. For m − , we get sin θ = 0.1175, θ = 6.7478◦ and ∆ θ = 8.7 ′ again.
N
(b) At the minimum, we found θ = 7.0364◦ or sin θ = 0.1225. Thus,
a sin θ (10−3 cm (0.1225)
λ= = = 6.125 × 10−5 cm = 612.5nm
m 2

1
If we had taken m− = (2 − 1/24) instead, sin θ = 0.1175 and
N
a sin θ (10−3 cm) (0.1175)
λ= = = 5.875 × 105 cm = 587.5nm
m 2
(c) The minimum angular separation is (∆ λ)m in = 612.5 nm − 600 nm = 12.5nm. So,
λ 600
Rm=2 = = = 48
∆ λm in 12.5
Using Eq. (12 11), R = m N = (2) (24) = 48.
65
λ a v 589.2935 a
12-4. R = m N = = = 987 ; N = . So, for m = 1, N = 987 and for m = 2, N = 494.
∆λ 0.597 m
12-5. The free spectral range is FSR = λ2 − λ1 = λ1/m. Let λ1 = 350 nm, crown ; λ1 = 180 nm, quartz.
(a) Then,
λ1 350 nm
λ2 = λ1 + = 350 nm + = 700 nm, crown
m 1
180 nm
λ2 = 180 nm + = 360 nm, quartz
1
(b) With a = 1/1200mm,
(1) (700 × 10−6)

θ = sin (m λ/a) = sin
−1 −1
= 57.14◦, crown
1/1200
(1) (360 × 10−6)

θ = sin−1 (m λ/a) = sin−1 = 25.59◦, quartz
1/1200
(c) FSR = λ1/m:
m = 1: FSR = 350/1 = 350 nm, crown; FSR = 180/1 = 180 nm, quartz
m = 2: FSR = 350/2 = 175 nm, crown; FSR = 180/2 = 90 nm, quartz
12-6. See Figure 12 14 that accompanies the statement of this problem in the text.
R (m = 3) = m N = (3) (16, 000 × 2.5) = 120, 000

R (m = 2) = m N = (2) (16, 000 × 2.5) = 80, 000
∆ λ = λ/R = 550 nm/80, 000 = 0.006875 nm = 0.069 Å
λ 5893
12-7. (a) R ≡ = = 982.16 ⇒ 983, m = R/N = 983/400 = 2.46; order = 3
∆λ 6
(b) Any width smaller than the light beam.
12-8. (a) We have m λ = a sin θ ≈ a ym/f or ym = m λ f /a. Then,
m = 0 : y0 = 0
(1) (546 × 10−6) (2000)
m = 1 : y1 = mm = 218.4mm
0.005
Thus ∆ y = y1 − y0 = 21.84cm , independent of N .
(b) The first missing order occurs at M = a/b = 5.

Thus, the number of maxima under the central peak is 2 (M − 1) + 1 or 2 M − 1 = 9. This is true for all 3
cases.

1
(c) For minima: m+ λ = a sin θ. Using zero order (m = 0), sin θ ≈ θ = λ/a N . So
N
Peak width = 2 θ = 2 λ/a N ; ∆ y = f ∆ θ = 2 f λ/a N

2 (200) (546 × 10−7)
N = 2: ∆ y = cm = 21.84 cm, N = 10: ∆ y = 4.368 cm, N = 15, 000: ∆ y = 0.0029 cm
(0.0005) (2)
66
12-9. (a) The resolution is R = λ/∆ λ = 3500 Å/0.02 Å = 175, 000 = m N . So for m = 2, N = 87, 500 grooves so that
there are 87, 500/10 = 8750 grooves/cm.
(b) At normal incidence,
−1 2 × 350 × 10 cm
−7

1 mλ 1
θB = sin−1 sin = 18.89◦
2 a 2 1/8750cm
(c) m 2 rad
(d) D = = = 22139
a cos θ (1/8750) cos 37.77 cm
θb rad 180 deg cm
θb
θ D = 22139 × × 10−7 = 0.1268 deg/nm
θb cm π rad nm
dλ 1
= = 7.88 nm/deg
θ = 2 θb = 37.77◦ dθ D
12-10. (a) The resolution is given by R = λ/∆λmin . Taking the worst case, or λ = 700 nm,
7000 Å
R= = 7000 = m N = (1)N ⇒ N = 7000
1 Å
Then, a = 2 cm/7000 grooves.
(b) The grating equation gives m λ = a sin θ ≈ a θ ≈ a y/f . Thus, m ∆λ = a ∆y/f . S0,
m f ∆λ (1) (50 cm) (1 × 10−8 cm)

∆y= = = 0.00175cm
a (2/7000)cm
mλ
12-11. (a) Using m λ = a (sin θi = sin θm) ⇒ sin θm = − sin θi,
a

(1) (400)
λ = 400 nm: θm = sin −1 − sin 30 = − 5.74◦
◦
1000

(1) (700)
λ = 700 nm: θm = sin−1 − sin 30◦ = + 11.54◦
1000
(b) W = 10 cm: R = m N = (1) (10, 000cm−1) (10 cm) = 100, 000
(c) The plate factor is d λ/d y = d λ/f d θ = 1/f D. The angular dispersion is,
tan (sin−1 m λ/a) tan (sin−1(550/1000))

D = tan θ/λ = = = 0.0011974 rad/nm
λ 550nm
So the plate factor is
dλ 1
= = 8.3516 nm/cm = 8.35 Å/mm
dy (100 cm) (0.00119737)nm−1
(d) Rcircle = Rgrating /2 = (1/2) (2 m) = 1 m
12-12. The plate factor P .F .,

1 d cos θ m
P .F . = = , since D = .
fD fm a cos θ
Thus,
1 gr cos θ
=
a cm m f (P .F .)
The answer depends on θ. In first order we suppose the P .F . is valid near θ = 0◦ so that cos θ ≈ 1. Then,
1 1 mm nm
≈ = 0.005 × 106 = 5000 grooves/cm
a (1) (100 cm) (2 nm/mm) nm − cm mm
67
12-13. (a) For the Littrow mount: m λ = 2 a sin θb, and so,
2 a sin θb
λ= = 2 (mm/300) sin 10◦ = 0.00116 mm = 1.16 µm.
m
(b) See the solution to problem 12 12. In this case,
a cosθ (1 mm/300) (cos 20.322◦) (106 nm/cm)
P .F . = = = 1.84 nm/mm
mf (1) (170 cm)(10 mm/cm)
(Here, sin θ = m λ/a = 2 sin θb = 2 sin 10◦ ⇒ cos θ = cos 20.322◦)
12-14. See Figure 12 16 that accompanies the statement of this problem in the text. Generally,
R = m N = m (W /a) = (1) (15 cm/a) or a = (15 cm)/R = (15 cm)/300, 000 = 5 × 10−5cm = 500nm
1 × 200

mλ
(a) Littrow: θb = sin−1 = sin−1 = 11.5◦
2a 2 × 500
−1 1 × 200

−1 m λ
(b) Normal: 2 θb = sin = sin = 23.578◦. So that, θb = 11.8◦
a 500
12-15. As given in the statement of the problem,

λ 488 × 10−6 mm 1
d= = = 2.8175 × 10−4 mm ⇒ = 3550 grooves/mm
2 sin θ 2 sin 60◦ d
If the emulsion is of refractive index n, then λ → λ0/n and d = λ0/(2 n sin θ). As n increases the grating con
stant increases.
λ 6563 6563
12-16. (a) R = = = m N or N = = 3647 grooves
∆ λm in 1.8 1.8 (1)
!
6300 Å 6300 Å

m λ 1 ×
(b) Littrow: θb = sin−1 or 22.2◦ = sin−1 ⇒ = 0.37784 ⇒ a = 8.337 × 10−5 cm.
2a 2a 2a
1
Thus, = 11995 grooves/cm ≈ 1200 grooves/mm.
a
3647gr
(c) Wm in = = 3.04mm
1200 gr/mm
12-17. (a) Littrow: m λ = 2 a sin θb,

2 a sin θb 2 (1 cm/80) sin 63◦ sin 63◦ cm
So m = = = .
λ λ 40 λ
λ = 400 × 10−7 cm, m = 557
λ = 700 × 10−7 cm, m = 318
W
(b) N = = (12 cm) (80 gr/cm) = 960 grooves
a
sin 63

1
(c) Now, R = m N . At λ = 550 nm, m = = 405. So R = (405) (960) = 388, 800.
40 550 × 10−7
λ 5500 Å
Then,(∆ λ)min = = = 0.014 Å
R 388, 800
m 405 rad 180 deg cm
(d) D = = cm = 71367 × × 10−7 = 0.41 deg/nm.
a cos θ cos 63◦ cm π rad nm
80
λm in 350 sin 63 350nm

1
(e) FSR = = , where m = = 636.4. So, FSR = = 0.55 nm = 5.5 Å.
m m 40 350 × 10−7 637
68
Chapter 13 Fresnel Diffraction
13-1. The far field is defined by

Area π d2 π(10−3m)2
L≫ = = = 1.4 m
λ 4λ 4 (546 × 10−9 m)
Thus L = 50 cm is the near field, L = 1 m is in the near field and L = 5 m is marginally in the far field.
13-2. See Figure 13 9 that accompanies the statement of the problem in the text. From Eq. (13 23),
R12 f1
fn = =
nλ n
Then, the maxima occur for odd values of n at,
(0.15 cm)2 f f
f1 = = 409 cm, f3 = 1 = 136 cm, f5 = 1 = 81.8 cm
(1) (5.5 × 10−5 cm) 3 5
The minima occur at even values of n,
f1 f f
f2 = = 204.5 cm, f4 = 1 = 102 cm, f6 = 1 = 68.1 cm
2 4 6
13-3. For a fixed distance f , we have the relation for the maxima and minima, f = r0 = Rn2 /nλ, with the maxima
occurring for odd n and the minima occurring for even n. Then the hole diameters are, then
√
Dn = 2 Rn = 2 n λ r0
p
First Max: n = 1: D1 = 2 (1) (5.893 × 10−5 cm) (150 cm) = 0.188cm
p
Second Max: n = 3: D3 = 2 (3) (5.893 × 10−5 cm) (150 cm) = 0.326cm
p
First Min: n = 2: D2 = 2 (2) (5.893 × 10−5 cm) (150 cm) = 0.266cm
p
Second Min: n = 4 D4 = 2 (4) (5.893 × 10−5 cm) (150 cm) = 0.376 cm
√ q
13-4. (a) R1 = n λ r0 = (1) 6 × 10−5 cm)(20 cm) = 0.0346 cm
(b) R1 = 1 cm; N = R12/λ r0 = 12/(6 × 10−5 × 20) = 833
(c) No minima are appear now. Maxima are

R12 (0.0346 cm)2 f f
f1 = r0 = = = 20 cm; f3 = 1 = 6.67 cm; f5 = 1 = 4 cm
(1)λ (6 × 10−5 cm) 3 5
13-5. See Figure 13 20 that accompanies this problem in the text for the geometry of the set up. Let d1 be the
distance from the source point to the indicated point on the wavefront in the aperture and d2 be the dis
tance from the indicated point on the wavefront in the aperture to the observation point P . For the nth
zone,
(d1 + d2) − (p + q) = nλ/2
From the geometry,
1/2
R2 1 Rn2 1 R2n
p
d1 = p2 + R2n = p 1 + 2n ≈p 1+ ; d 2 ≈ q 1 +
p 2 p2 2 q2
So the condition becomes,

1 1 1 1 1
(d1 + d2) − (p + q) = R2n + ≡ Rn2 = nλ/2
2 p q 2 L
√
Rn = n λ L
69
13-6. Using the set up of problem 13 5,
1 1 1 1 1
= + = + ⇒ L = (20/3) cm
L p q 10 cm 20 cm
√ p
(a) R1 = n λ L = (1) (20/3) (6 × 10−5) cm = 0.2 mm
R21 (1)2
(b) N = = = 2500.
λ L (20/3) (6 × 10−5)
13-7. Using the results of problem 13 5,

1 1 1 1 1
= + = + ⇒ L = 25 cm
L p q 50 cm 50 cm
√ R2
Rn = n L λ = n = n
Lλ
(a) Then,
2
(0.05 cm)
ninner = =2
(25 cm) (5 × 10−5 cm)
2
(0.0935 cm)
nouter = =7
(25 cm) (5 × 10−5 cm)
So, zones 3, 4, 5, 6, and 7 contribute. Every other mode essentially cancels the preceding zone so only one
zone contributes. For the unobstructed wave front, Iu ∝ (a1/2)2 and here I ∝ a21 assuming a1 ≈ a2 ≈ ...a7. So,
I/Iu = 4.
(b) In this case,

2
(0.1 cm)
nouter = =8
(25 cm) (5 × 10−5 cm)
In this case zones 3, 4, 5, 6, 7, and 8 contribute and the contributions nearly cancel in pairs giving I ≈ 0.
(c) From above, 5 for part (a) and 6 for part (b).
13-8. From Eq. (13 15), the surface area of the nth Fresnel zone is
2 !
π r0′ r02 λ

λ
Sn = + (2 n − 1)
r0 + r0′ r0 2 r0
So that, with λ/r0 = (5 × 10−5/50) = 10−6

S25 − S1 48 (λ/2 r0)2
= = 1.2 × 10−5 = 0.0012%
S1 (λ/r0) + (λ/2r0)2
13-9. (a) Before reduction R1 = 11.25 cm.

√ p
The required radius is R1 = n λ f1 = (632.8 × 10−5) (200) = 0.1125 cm.
It must be reduced by a factor of 100.

√ √
(b) RN = N R1 = 20 (11.25 cm) = 50.31 cm.
√
13-10. See Figure 13 21 that accompanies the problem in the text. Using Rn = n r0 λ ,
√ p
D2 = 2 R2 = 2 2 r0 λ = 2 2 (25) (5.5 × 10−5) cm = 0.105 cm
√ p
D4 = 2 R2 = 2 4 r0 λ = 2 4(25) (5.5 × 10−5) cm = 0.148 cm
√ p
D6 = 2 R2 = 2 6 r0 λ = 2 6(25) (5.5 × 10−5) cm = 0.182 cm
70
13-11. The zone areas are given by Eq. (13 15) which goes to Eq. (13 16) in the limit λ/r0 = λ/x ≪ 1. Eq. (13 16)
is independent of n, so the areas are approximately constant. For a plane wave, r0 → ∞, thus from Eq. (13
16),
π r0′ r0 πλ
Sn = πλ = Lim = π λ r0′ = π λ x
r0 + r0′ r0 →∞ 1 1
+ r′
r 0 0
√ p
13-12. The radius is given by R4 = 4 r0 λ = (4 )(2) (485 × 10−9) m = 0.00197 m = 1.97 mm. Since 4 zones con
tribute and the zones nearly cancel in pairs, the irradiance is nearly zero.
13-13. Since p → ∞,
1 1 1 1
= + = ⇒L= q
L p q q
Also ∆z = w. Using Eq. (13 28),
r
2 2 w2 2 (0.0005 m)2
∆v = ∆z ⇒ L = = = 0.148 m = 14.8 cm
Lλ (∆v)λ (2.5)2 (540 × 10−9 m)
13-14. See Figure 13 22 that accompanies the statement of this problem in the text. Also it may be helpful to
examine Figure 13 15 in the body of the text. The parameter L is,
1 1 1 1 1
= + = + ⇒ L = 10 cm
L p q 30 cm 15 cm
(a) For an on axis point, z2 = w/2 = 2.5 × 10−4 m and z1 = − w/2 = − 2.5 × 10−4 m. Then
r r
2 2
v2 = z2 = (2.5 × 10−4) = 1.694
Lλ (0.1) (435.8 × 10−9)
r r
2 2
v1 = z1 = ( − 2.5 × 10−4) = − 1.694
Lλ (0.1) (435.8 × 10−9)
Then, interpolating in Table 13 1,
C(v2) ≈ 0.3265, S(v2) = 0.5549

C(v1) = − 0.3265, S(v1) = − 0.5549
Using Eq. (13 35),

n o
I = I0 [C(v2) − C(v1)]2 + [S(v2) − S(v1)]2
I = I0 (2 × 0.3265)2 + (2 × 0.5549)2 = 1.658 I0 = 1.658 Iu/2 = 0.829 Iu

(b) The edge of the geometrical shadow on the screen is at y = 0.375 mm. The axis from S to P for this
point brushes the top edge of the aperture at z2 = 0. Then the bottom edge of the aperture is located at
z = − w = − 5 × 10−4 m. Then,
r r
2 2
v2 = 0, v1 = z1 = ( − 5 × 10−4) = − 3.388
Lλ (0.1) (435.8 × 10−9)
C(v2) = 0, S(v2) = 0; C(v1) = − 0.42746, S(v1) = − 0.49328
n o
2 2
I = I0 [C(v2) − C(v1)] + [S(v2) − S(v1)]
I = I0 0.42762 + 0.493282) = 0.4262I0 = 0.4262 (Iu/2) = 0.2131 Iu
71
13-15. Refer to Figure 13 23 that accompanies the statement of the problem in the text and Figure 13 4a in the
body of the text. For both positions,
1 1 1
= + ⇒ L = 40 cm = 0.4 m
L 60 cm 120 cm
(a) Let z ′ be the coordinate of the point O ′ in the aperture plane along the straight line from S to the
observation point P . (See Figure 13 4a). In the case at hand
p 60
z′ = y= ( − 2 mm) = − 6.667 × 10−4 m.
p+q 60 + 120
The values of the z1 and z2 that mark the edges of the unobstructed regions in the aperture plane are to be
measured relative to this point so that z2 = ∞ and z1=+6.667 × 10−4 m
r r
2 2
6.667 × 10−4 = 1.942

v2 = ∞, v1 = z1 =
λL (589.3 × 10 )(0.4)
−9
C(v2) = 0.5, S(v2) = 0.5, C(v1) = 0.4338, S(v1) = 0.3608

n o
I = I0 [C(v2) − C(v1)]2 + [S(v2) − S(v1)]2
I = I0 0.5 − 0.4338)2 + (0.5 − 0.3608)2 = 0.02376 I0 = 0.02376 (Iu/2) = 0.0119 Iu

p 60
(b) In this case z ′ = y= (1 mm) = 3.333 × 10−4 m. Then, z2 = ∞ and z1 = − 3.333 × 10−4 m
p+q 60 + 120
r r
2 2
− 3.333 × 10−4 = − 0.9709

v2 = ∞, v1 = z1 =
λL (589.3 × 10 )(0.4)
−9
C(v2) = 0.5, S(v2) = 0.5, C(v1) = − 0.7755, S(v1) = − 0.4097

n o
I = I0 [C(v2) − C(v1)]2 + [S(v2) − S(v1)]2
I = I0 0.5 + 0.7755)2 + (0.5 + 0.4097)2 = 2.454 I0 = 2.454 (Iu/2) = 1.23 Iu

13-16. The parameter L is given by

1 1 1
= + ⇒ L = 20 cm = 0.2 m
L 30 cm 60 cm
(a) The unobstructed region in the upper half of the aperture plane relative to the point in the center of
the wire on the line connecting the source and the observation point is bounded by the points zu2 = ∞ and
zu1 = 7.5 × 10−4 m. The unobstructed region in the lower half of the aperture plane is bounded by the
points zl2 = − 7.5 × 10−4 m and zl1 = − ∞. Then,
r r
2 2
7.5 × 10−4 = 3.2094

vu2 = ∞, vu1 = z1 =
λL (546.1 × 10 )(0.2)
−9
C(vu2) = 0.5, S(vu2) = 0.5, C(vu1) = 0.4607, S(vu1) = 0.5863

vl2 = − 3.2094, vl1 = − ∞
C(vl2) = − 0.4607, S(vl2) = − 0.5863, C(vl1) = − 0.5, S(vl1) = − 0.5
n o
2 2
I = I0 [C(vl2) − C(vl1) + C(vu2) − C(vu1)] + [S(vl2) − S(vl1) + S(vu2) − S(vu1)]
nh o
I = I0 − 0.4607 − ( − 0.5) + 0.5 − 0.4607]2 + [ − 0.5863 − ( − 0.5) + 0.5 − 0.5863]2 = 0.03599I0
I = 0.03599 (Iu/2) = 0.018 Iu
Part (b) is on the next page.
72
13-16. (b) The point on the screen that is at the edge of the geometrical shadow is at position
p+q 90
y= zwire top = (0.75 mm) = 2.25 mm
p 30
Following the procedure from part (a):
zu1 = 0, zu2 = ∞, zl1 = − ∞, zl2 = − 1.5 × 10−3 m
r
2
− 1.5 × 10−3 = − 6.42

vu1 = 0, vu2 = ∞, vl1 = − ∞, vl2 =
(546.1 × 10 )(0.2)
−9
C(vu2) = 0.5, S(vu2) = 0.5, C(vu1) = 0, S(vu1) = 0

n C(vl2) = − 0.5414 , S(vl2) = − 0.5138, C(vl1) = − 0.5, S(vl1) = − 0.5 o
I = I0 [C(vl2) − C(vl1) + C(vu2) − C(vu1)]2 + [S(vl2) − S(vl1) + S(vu2) − S(vu1)]2
n o I
I = I0 [ − 0.5414 − ( − 0.5) + 0.5 − 0]2 + [ − 0.5138 − ( − 0.5) + 0.5 − 0]2 = 0.4467I0 = 0.4467 u = 0.223 Iu
2
13-17. Refer to the routine used to solve problem 13 15b. Borrowing from that solution,
n o n o
I = I0 [C(v2) − C(v1)]2 + [S(v2) − S(v1)]2 = I0 [0.5 − C(v1)]2 + [0.5 − S(v1)]2
From Table 13 1 the second maximum and minimum occur for,
Max: v1 = − 2.3, C(v1) = − 0.626, S(v1) = − 0.5531
I2n d m ax = 2.378 I0 = 2.378 (Iu/2) = 1.19 Iu
Min: v1 = − 2.75, C(v1) = − 0.4300, S(v1) = − 0.4222
I2n d min = 1.715 I0 = 1.715 (Iu/2) = 0.86 Iu
1 1 1
13-18. The parameter L is given by = + ⇒ L = 1.2 m. The unobstructed region in the upper half of the
L 2m 3m
aperture plane relative to the point in the center of the wire on the line connecting the source and the
observation point is bounded by the points zu2 = ∞ and zu1 = 1.85 × 10−4 m. The unobstructed region in
the lower half of the aperture plane is bounded by the points zl2 = − 1.85 × 10−4 m and zl1 = − ∞. Then,
r r
2 2
1.85 × 10−4 = 0.3009

vu2 = ∞, vu1 = z1 =
λL (630 × 10 )(1.2)
−9
C(vu2) = 0.5, S(vu2) = 0.5, C(vu1) = 0.3003, S(vu1) = 0.0143

vl2 = − 3.2094, vl1 = − ∞
n C(v l2) = − 0.3003, S(v l2) = − 0.0143, C(vl1) = − 0.5, S(vl1) = − 0.5 o
I = I0 [C(vl2) − C(vl1) + C(vu2) − C(vu1)]2 + [S(vl2) − S(vl1) + S(vu2) − S(vu1)]2
nh o
I = I0 − 0.3003 − ( − 0.5) + 0.5 − 0.3003]2 + [ − 0.0143 − ( − 0.5) + 0.5 − 0.0143]2 = 1.103I0
I = 1.103 (Iu/2) = 0.55Iu
13-19. See Figure 13 25 that accompanies the statement of the problem in the text. The parameter L is given by
1 1 1
= + ⇒ L = 5 cm
L 10 cm 10 cm
Each slit contributes the same electric field at the on axis point and so the phasors associated with each slit
will simple add. Then for the upper slit,
z1 = (0.25/2 − 0.02) cm = 0.0105 cm, z2 = (0.25/2 + 0.02) cm = 0.0145 cm
r r
2 2
v1 = z1 = (0.0105 ) = 0.8987, v2 = 1.2411
λL 5 (5.46 × 10−5)
n C(v1) = 0.76426, S(v1) = 0.33862, C(v2) = o 0.68384, S(v2) = 0.64925
I = I0 [2 (C(v2) − C(v1))]2 + [2 (S(v2) − S(v1))]2 = 0.412 I0 = 0.412 Iu/2 = 0.206 Iu
73
13-20. The parameter L is given by
1 1 1
= + ⇒ L = 12.5 cm
L 25 cm 25 cm
(a) The far field condition is
p, q ≫ b2/λ = (0.075 cm)2/(4.358 × 10−5 cm = 129 cm

Since for the case described in the problem statement p = q = 25cm, far filed approximations are not justi
fied.
(b) Using the far field approximations for comparison:
m λ = b sin θ (take m = 1)
sin θ = λ/b = (4.358 × 10−5)/0.075 ⇒ θ = 0.03329◦
y = q tan θ = (25 cm) tan (0.00329◦) = 0.0145 cm
(c) See Figure 13 15a for the geometry. In the case at hand, y = 0.0145 cm so that the point O ′ in the aper
ture plane and connecting the source to the observation point is positioned at,
p 1
z′ = y = y = 0.00725 cm
p+ q 2
above the center of the aperture. Relative to this point the edges of the slit are at positions
0.075 0.075

z1 = − + 0.00725 cm = − 0.04475 cm, z2 = − 0.00725 cm = 0.03025 cm
2 2
Then,
r r
2 2
v1 = z1 = ( − 0.04475) = − 2.7115
Lλ (12.5) (4.358 × 10−5)
r r
2 2
v2 = z1 = (0.03025) = 1.8329
Lλ (12.5) (4.358 × 10−5)
C(v1) = − 0.401, S(v1) = − 0.446; C(v2) = 0.354, S(v1) = 0.425
n o
2 2
I = I0 [C(v2) − C(v1)] + [S(v2) − S(v1)]
n o
I = I0 (0.354 + 0.401)2 + (0.425 + 0.446)2 = 1.33I0 = 1.33(Iu/2) = 0.665 Iu
13-21. According to Babinet’s principle, the diffraction pattern is complementary to that due to circular apertures
of the same size as the particles, in an otherwise opaque screen. The halo is the complement of the Airy
disc. Taking λ = 550 nm,
1.22 λ 1.22 (550 × 10−9 m)
a= = ≈ 19 µm
sin θ sin (2◦)
74
Chapter 14 Matrix Treatment of Polarization
14-1. If the light is polarized along the transmission axis oriented at angle θ the light is transmitted unchanged
so that,
a b cos θ cos θ

=
c d sin θ sin θ
If light is polarized perpendicular to the transmission axis no light is transmitted,
a b cos (θ − 90◦) sin θ

a b 0
= =
c d sin (θ − 90◦) c d − cos θ 0
These two matrix equations can be recast as four algebraic equations,
a cos θ + b sin θ = cos θ (1)

c cos θ + d sin θ = sin θ (2)
a sin θ − b cos θ = 0 (3)
c sin θ − d cos θ = 0 (4)
An approach that can be used is multiply Eq. (1) by sin θ and Eq. (3) by cos θ and then subtracting the
results. Then,
b sin2 θ + cos2 θ) = cos θ sin θ ⇒ b = cos θ sin θ
Proceeding in the same manner with Eqs. (2) and (4) gives,
c = sin θ cos θ
Then using these relations for b and c in (3) and (4) leads to,
a = cos2 θ, d = sin2 θ
So that,
sin2 θ sin θ cos θ

a b
=
c d sin θ cos θ cos2 θ
E0xeiϕx i(kz −ωt)

14-2. In general Ẽ = E0xeiϕxxx + E0y eiϕy ŷ ei(kz −ωt) =

e = Ẽ0
E0y eiϕy

i(kz −ωt) 1 1 1
(a) Ẽ = [E0x̂ − E0 ŷ]e ⇒ Ẽ0 = E0 ⇒√ . Linearly polarized at − 45◦.
−1 2 −1

1 1 1
(b) Ẽ = [E0x̂ + E0 ŷ]ei(kz −ωt) ⇒ Ẽ0 = E0 ⇒√ . Linearly polarized at 45◦.
1 2 1
 
1

h i 1 1
(c) Ẽ = E0x̂ + E0e−iπ/4 ŷ ei(kz −ωt) ⇒ Ẽ0 = E0 −iπ/4 ⇒ √   1
. Then,
e 2 √ (1 − i)

2
2E0xE0ycos ε
tan 2α = 2 2 → ∞ ⇒ 2 α = 90◦, α = 45◦
E0x − E0y
Right elliptically polarized at 45◦.

h i 1 1 1
(d) Ẽ = E0x̂ + E0eiπ/2 ŷ ei(kz −ωt) ⇒ Ẽ0 = E0 iπ/2 ⇒ √ . Left circularly polarized.
e 2 i
75
E0xeiϕx i(kz −ωt)

14-3. In general Ẽ = E0xe x̂ + E0y e yy e
i(kz −ωt)
iϕx iϕ y
= Ẽ0ei(kz −ωt)

= e
E0y eiϕy

1
(a) Ẽ = (2 E0x̂ + 0ŷ )ei(kz −ωt) ⇒ Ẽ0 = 2 E0 . Linearly polarized along the x direction. Velocity is in the
0
√
+ z direction. The amplitude is A = 2 E0 12 + 02 = 2 E0.

3
(b) Ẽ = (3E0x̂ + 4E0 ŷ )ei(kz −ωt) ⇒ Ẽ0 = E0 . The polarization direction makes the angle α with the x
4
axis where,
α = tan−1(4/3) = 53◦
√
The wave is traveling in the + z direction with amplitude A = 32 + 42 E0 = 5 E0.

1
(c) Ẽ = 5 E0(x̂ − iŷ )ei(kz+ωt) ⇒ Ẽ0 = 5E0 . The propagation is in the +z direction. The wave is right
−i
circularly polarized with amplitude. The electric field vector traces out a circle of radius 5 E0.

i(kz −ωt) 1
14-4. (a) Ẽ1 = E01 (x̂ − ŷ )e ⇒ Ẽ01 = 2 E01 . This is linearly polarized along − 45◦
−1
√ √

3 1
Ẽ2 = E02 3 x̂ + ŷ ei(kz −ωt) ⇒ Ẽ02 = E02 . This is linearly polarized along α = tan−1 √ = 30◦
1 3
The angle between the two is 75◦. √
√ √ √ 3−1
(b) Ẽ01 · Ẽ02 = E01 E02 ( 3 − 1) = ( 2 E01 ) ( 3 + 1 E02 ) cos(θ12 ) ⇒ cos θ12 = √ ⇒ θ12 = 75◦
2
2 2
14-5. Since we have not determined a matrix for a HWP with its FA at an arbitrary at 45◦, we do the problem in
two steps. First consider the effect of passing linearly polarized light at − 45◦ through a HWP with FA ver
tical,
1 0 1 1
=
0 −1 −1 1
HW P, FA vert LP at −45 ◦
The output is linearly polarized at 45◦. Thus the first step rotates the polarization through an angle of 90◦.
If the linearly polarized light is originally vertical it exits the HWP horizontal. Then the LP and QWP act
as,
1 0 1 1 1 1
=
0 −i 1 1 0 −i
QW P LP
The output light has right circular polarization.
14-6. (a) The wave propagates in the +x direction so the phase factor is ei(kx−ωt). The angle α = 30◦ so that
1 E √
tan α = 30◦ = √ = 0z . Then Ẽ = E0( 3 ŷ + ẑ )ei(kx−ωt).
3 E0y
(b) The phase factor is ei(ky −ωt). Also,

A 2
with A = 2 B ⇒ E0
−iB −i
So, Ẽ = E0(2 ẑ − ix̂)ei(ky −ωt)
|k| |k|
(c) The propagation vector is k = √ (x̂ + ŷ ). Then k · r = √ (x + y).
2 2
h √
i
i |k|(x+ y)/ 2 −ωt
Then Ẽ = E0 ẑ e .
76
14-7. The general form is,
A
Ẽ0 =
B+iC
(a) Linear polarization: C = 0. Phase differences must be multiples of m π.

(b) Elliptically polarized: major axis along a coordinate axis: B = 0. The phase difference should be
(m + 1/2) π.
(c) Circularly polarized: B = 0 and C = ± A, with phase difference (m + 1/2) π.
14-8. The form of the ellipse is given as,

2 2
Ex Ey Ex Ey
+ −2 cos ε = sin2 ε
E0x E0y E0x E0y
This has the form of a quadratic,

E y2 + b E y + c = 0
with,
Ex2

E0y
b=−2 (cos ε) Ex and 2
c = E0y 2
− sin2 ε
E0x E0x
√

A
As shown in the text, when Ẽ0 = , E0x = A, E0y = B 2 + C 2 , and ε = tan−1(C/B)
B +iC
Solving the quadratic gives,
√
−b± b2 − 4 c
Ey =
2
So knowing the form of the Jones vector allows for the construction E y for given Ex. I’ll illustrate this pro
cess using the free computer algebra system Maxima:
(%i1) (A:3,B:2,C:1)$
(%i2) epsilon:atan(C/B)$
(%i3) (E0x:A,E0y:sqrt(B^2+C^2))$
(%i4) for j from -30 thru 30 do
(Ex[j]:0.1*j,b[j]:-2*E0y/E0x*cos(epsilon)*Ex[j],c[j]:E0y^2*(Ex[j]^2/E0x^2-(sin(epsilon))^2),
Ey1[j]:-b[j]/2-1/2*sqrt(b[j]^2-4*c[j]),Ey2[j]:-b[j]/2+1/2*sqrt(b[j]^2-4*c[j]))$
(%i5) for j from 0 thru 0 do ldisplay(Ex[j],Ey1[j],Ey2[j])
(%t5) Ex0 = 0
(%t6) Ey10 = − 1
(%t7) Ey20 = 1
Note that I have displayed the first set of points values. The loops could be extended to form any number of the values. The
two E y values correspond to the two roots of the quadratic. The list so generated can be plotted to form the ellipse. This is
done below.
(%i7) (Eylist:listarray(Ey1),Exlist:listarray(Ex),Eylist2:listarray(Ey2))$
(%i8) plot2d([[discrete,Exlist,Eylist],[discrete,Exlist,Eylist2]],[gnuplot_term,ps],[gnuplot_ps_term_command,
"set term postscript eps monochrome"],[gnuplot_out_file, "ellipse.eps"])
25
Ey discrete1
discrete2
2
15
05
Ex
05
15
25
3 2 1 0 1 2 3
77

3i 3
14-9. (a) →i Linear polarization at α = tan−1(1/3) = 18.4◦
i 1

i 1
(b) →i Right circular polarization, A = 1
1 −i

4i 4
(c) →i Right elliptical polarization, axes 5 along y, 4 along x
5 − 5i

5
(d) Linear polarization along x, A = 5
0

2 1
(e) →2 Left circular polarization. A = 2
2i i
√ √

2
(f) Linear polarization at α = tan−1(3/2) = 56.3◦, A = 22 + 32 = 13
3
√

2 1
(g) =2 Left elliptical polarization, E0x = 2, E0y = 2 32 + 42 = 10, ε = tan−1(4/3) = 53.1◦
6+8i 3+4i
(2)(1)(5) cos 53.1◦
tan 2 α = ⇒ α = − 7.02◦
12 − 52
14-10. Using the Jones formalism, √ √

1 0 3 3
=
0 −i 1 −i
QW P LP, 30 ◦
√
The output light is has right elliptical polarization with E0x/E0y = 3.
14-11. Using the Jones formalism,

cos α cos α cos( − α)

1 0
= =
0 −1 sin α − sin α sin( − α)
HW P LP,α LP, −α
The input light is rotated by 2 α.
14-12. Light from the 45◦ linear polarizer passes through the QWP twice, before and after reflection. Thus it
passes through (equivalently) a HWP that rotates the direction of polarization by 2 × 45◦ = 90◦. (See
problem 14 11.) It then returns to the linear polarizer perpendicular to the transmission axis. As a result
no light passes back through the polarizer after being reflected.
14-13. See Figure 14 13 that accompanies the statement of this problem in the text. Using the Jones formalism,

1 0 1 1 1 1 1 1 1
= = Right circular polarization
0 −i 1 1 0 −i −i 0 −i
QW P LP
SA, hor at 45◦

0 0 1 0 1 0 1 0 1 0 0 1 0 1 0
= = No light
0 1 0 −1 0 0 0 i 1 0 −1 0 0 1 0
LP HWP LP QWP LP
TA vert TA h o r FA hor at 45 ◦

√
cos 30◦

0 0 1 0 0 0 3 0 0
= = =−i
0 1 0 −i sin 30◦ 0 1 −i −i 1
LP QWP LP (a) Right (b)LP
TA ver t SA hor at 30 ◦ Ellipt vert
78
14-16. (a) In all cases the light is linearly polarized light at ± 45◦.

1 0 1 1
= : Linear polarization at at + 45◦
0 −i i 1
SA −x LC P

1 0 1 1
= : Linear polarization at at − 45◦
0 −i −i 1
SA −x RCP

1 0 1 1
= : Linear polarization at at − 45◦
0 i i 1
SA − y LC P

1 0 1 1
= : Linear polarization at at + 45◦
0 i −i 1
SA − y RCP
(b) For an eighth wave plate: |ε y − εx | = 2 π/8 = π/4. Then the Jones matrix is,

1 0
M1/8 =
0 e±iπ/4
with the ± for εy > εx or ε y < εx. Then,

1 0 1 1 1 1 1
= = i(π/2± π/4) = i(3π/4) or i(π/4)
0 e±iπ/4 i i e±iπ/4 e e e
LC P

1 0 1 1 1 1 1
= = −i(π/2± π/4) = −i(3π/4) or −i(π/4)
0 e±iπ/4 −i − i e±iπ/4 e e e
RCP
Using Euler’s theorem the output light can be expressed as,

 

A

√ 1 √
= ± 2 ± i 2 
B +iC
2
√
This is of the form of elliptical polarization with A = 1, and |B | = |C | = 2 /2.
14-17. Consider the action of the matrix on a general Jones vector,

1 i A A+iB −C 1
= = (A − C + i B) : Right circular polarization
−i 1 B+iC −iA+B +iC −i
For a left circular polarizer try,

1 −i A A−iB +C 1
= = (A + C − i B) : Left circular polarization
i 1 B +iC iA+B +iC +i
14-18. Note that,
cos α cos α + 1

1 A
+ = =
±i sin α sin α ± i B ±iC
C ircular Linear Elliptical
79
14-19. In general, for elliptical polarization, E0x does not equal E0y and ε does not equal 0,
E y = E0y cos(k z − ω t) = E0y cos α
Ex = E0x cos(k z − ω t + ε) = E0x cos (α + ε)
Eliminate z and t through α:
Ey
= cos α
E0y
Ex E
= cos (α + ε) = cos α cos ε − sin α sin ε = y cos ε − 1 − (E y/E0y)2 sin ε
p
E0x E
p 0y
(Ex/E0x) − (E y/E0y) cos ε = − 1 − (E y/E0y)2 sin ε
Squaring gives,
(Ex/E0x)2 + (E y/E0y)2 cos2 ε − 2 (ExE y)cos ε/(E0xE0y) = (1 − (E y/E0y)2 sin2 ε
2 2
Ex Ey E E
cos2 ε + sin2 ε − 2 x y cos ε = sin2 ε

+
E0x E0y E0x E0y
2 2
Ex Ey E E
+ − 2 x y cos ε = sin2 ε
E0x E0y E0x E0y
14-20. (a) Putting the Jones vector into standard form,

A 2 2 2 √
= = =
B +iC 3eiπ/3 3cos (π/3) + i 3 sin(π/3) 3/2 + i 3 3 /2
√
So, A = 2, B = 3/2, and C = 3 3 /2. The normalized standard form is then

1 A 1 2 √ 1 Eox
√ =√ =√ iε
A2 + B 2 + C 2 B + i C 13 3/2 + i 3 3 /2 13 E0ye
This corresponds to elliptical polarization with
√
E0x = A = 2, E0y = B 2 + C 2 = 3, ε = tan−1(C/B) = π/3
Then,
E0x E0y cos ε 2 (2) (3) cos (π/3)
tan 2 α = 2 2 = ⇒ α = − 25.07◦
E0x − E0y 22 − 32
(b) The modification by the rotator is given by the Jones formalism with β = 30◦,
"√ # " √ √ #
cos β − sin β

2 3 /2 −√ 1/2 2 √ ( 3 − 3/4)
√ − i (3 3 /4)
= =
sin β cos β 3eiπ/3 1/2 3 /2 3/2 + i 3 3 /2 (1 + 3 3 /4) + i (9/4)
In general one can write,

a2 + b2

a−ib
(a + i b) =
c+id (a c − b d) + i (b c + a d)
√ √ √
For the case at hand, a = 3 − 3/4), b = 3 3 /4, c = 1 + 3 3 /4, and d = 9/4. So, the output Jones vector
can be written as,
" √ √ #
2.65192

( 3 − 3/4)
√ − i (3 3 /4) A
= =
(1 + 3 3 /4) + i (9/4) − 0.66506 + i 5.19615 B +iC
√
So, E0x = A = 2.652, E0y = B 2 + C 2 = 5.239, ε = tan−1(C/B) = 97.294◦. Then
E0x E0y cos ε 2(2.65) (5.24) cos (97.3◦)
tan 2 α = = ⇒ α = 4.903◦
2
E0x 2
− E0y (2.65)2 − 5.242
The elliptical polarization is thus rotated by (4.90 ◦ − ( − 25.07◦))=+30◦, as expected.
80
14-21. According to Eq. (14 12),
2 2
Ex Ey ExE y
+ −2 cos ε = sin2 ε
E0x E0y E0x E0y
(a) For ε = π/2 ⇒ cos ε = 0, sin ε = 1. Then,

2 2
Ex Ey
+ =1
E0x E0y
This is an ellipse with semi axes of E0x and E0y aligned with the coordinate axes.
2 2
(b) If E0x = E0y = E02,
Ex2 + E y2 − 2 Ex E y cos ε = E02 sin 2 ε
This is an ellipse with principal axes at 45◦ to the coordinate axes since,
E0x E0y cos ε 2 E02 cos ε

tan 2α = 2 2 = 2 = ∞ ⇒ 2 α = 90◦ ⇒ α = 45◦
E0x − E0y E0 − E02
(c) If the conditions of both (a) and (b) hold, the equation for the ellipse reduces to
Ex2 + E y2 = E02
This is the equation for a circle centered on the origin of radius E0.
(d) For ε = 0, cos ε = 1, sin ε = 0. Then,

2 2
Ex Ey
E xE y
+ =0 −2
E0x E0y
E0x E0y
2
Ex E E0y
− y = 0 ⇒ Ey = Ex
E0x E0y E0x
This is an equation for a straight line of slope E0y/E0x.
14-22. Take the fast axis of the QWP to be along the x axis. Then the light emerging from the first polarizer has
the Jones vector,
cos θ

sin θ
The second polarizer (sometimes called the analyzer ) has its transmission axis at α = 90◦ + θ. Then using
the general Jones matrix for a linear polarizer and the relations cos α = − sin θ and sin α = cos θ.
cos2α sin α cos α sin2θ − sin θ cos θ

=
sin α cos α sin2 α − sin θ cos θ cos2 α
Then the output light has the Jones vector,

sin2θ − sin θ cos θ 1 cos θ sin θ

0
= (1 − i) sin θ cos θ
− sin θ cos θ cos2 α 0 i sin θ cos θ
Analyzer QW P Am p =1 A mp =1
FA hor
The amplitude of the output vector (relative to the input vector) is (1 − i) sin θ cos θ. Then, with the input
irradiance being I0, the output irradiance is,
I = I0 sin2 θ cos2 θ |(1 − i)|2 = 2 sin2 θ cos2 θ I0
81
Chapter 15 production of polarized light
15-1. The first polarizer blocks 1/2 of the unpolarized incident light. Then applying Malus’ law for the last two
polarizers,
1
I1 = I0
2
1
I2 = I1 cos2(30◦) = I0 cos2 30◦
2
1
I3 = I2 cos2(60◦ − 30◦) = I0 cos4 30◦ = 0.2815 I0 ⇒ 28.15% of I0
2
n2
15-2. The polarizing angle is given by the relation, tan θ p = . So for nair = 1 and ndiam = 2.42
n1

nair 1
Internal reflection: θ p = tan−1 = tan −1
= 22.5◦
ndiam 2.42
2.42

ndiam
External reflection: θ p = tan−1 = tan−1 = 67.5◦
nair 1
15-3. (a) See Figure 15 22 that accompanies the statement of the problem in the text. Let the amplitude of the
unpolarized incident field component parallel to the transmission axis TA1 be E0p and let the amplitude of
2 2
the field component perpendicular to TA1 be E0 n with E0p + E0n = E02 and E0p
2 2
= E0n = E02/2. Then:
Polarizer 1:
√ √
E1p = α E0p through TA1 since the energy fraction is α. Similarly, E1n = β E0n through the extinction
axis EA1.
Polarizer 2: Now each component E1p and E1n has a component along both the transmission axis
TA2 and extinction axis EA2 of the second polarizer. Then
√ √
E2p = α (E1p cos θ) thru TA2 E2n = α (E1n sin θ) thru TA2
√ √ √ √
or E2p = α ( α E0p) cos θ thru TA2 E2n = 2 ( β E0n) sin θ thru TA2
√ √
E2p = β (E1p sin θ) thru EA2 E2n = β (E1n cos θ) thru EA2
√ √ √ √
E2p = β ( α E0p) sin θ thru EA2 E2n = β ( β E0n) cos θ thru EA2
I p = ε0 c/2)E p2 = ε0 c/2)[E2p (TA)]2 + [E2p (EA)]2 In = ε0 c/2)En2 = ε0 c/2)[E2n (TA)]2
2 2 2 2 2
I p = ε0 c/2) α E0p cos θ + α β E0P sin2 θ sin2 θ + β 2 E0n cos2 θ

In = ε0 c/2) α β E0n
2 2
I p = ε0 c/2)E0p (α2 cos2 θ + α β sin2 θ) In = ε0 c/2)E0n (α β sin2 θ + β 2 cos2 θ)
2 2
Using E0p = E0n = E02/2 one finds,
ε c E 2 1
I = I p + In = 0 0
(α + β) cos2 θ + 2α β sin2 θ = I0 (α + β) cos2 θ + 2α β sin2 θ

2 2 2
I = I0 (1/2)(α + β) cos2 θ + α β sin2 θ

1
In the ideal case, β = 0 and α = 1 so that I = I0 cos2 θ.
2
(b) With α = 0.95, β = 0.05, one finds
ideal actual 

θ
0◦ 0.5 0.4525 


30◦ 0.375 0.351 fractions of I0
45◦ 0.250 0.250 


90◦ 0 0.0475

82
λ λ 632.8 × 1 −−7 cm
15-4. = t (∆n) or t = = = 0.063mm
2 2 ∆n 2 (1.599 − 1.594)
15-5. See the answers in the back of the text.
15-6. (a) With the flat plate in the position of maximum retardation,
2λ 2 (546.1 × 10−6 mm
∆max = 2 λ = (∆n) (d2 − d1) or d2 − d1 = = = 0.12mm
∆n 1.555 − 1.546
λ 546.1 × 10−6 mm
(b) For circularly polarized light, ∆ = λ/4, so d2 − d1 = = = 0.015mm.
4 ∆n 4 (1.555 − 1.546)
15-7. (a) The sketch is essentially the same as Figure 15 22 that accompanies problem 15 3 in the text. Then,
1
I1 = I0/2 E2 = E1 cos θ ⇒ I2 = I1 cos2 θ = I0 cos2 θ
2
1 1
(b) In this case, I2 = I0 cos2 θ = I0 cos2 (90◦ − 0◦) = 0.
2 2
(c) The angle between successive polarizers is always 15◦ so,
1 2 ◦ 6
I7 = I0 cos (15 ) = 0.33 I0
2
So 33% of the incident light is transmitted.
15-8. The angular offset between successive polarizers is θ = 90◦/N . Applying Malus’ law N times in succession,
N 2N
IT = I0 cos2 θ = I0 [cos (90◦/N )] = 0.9 I0
2N
[cos (90◦/N )] = 0.9
A numerical solution indicates that N is between 23 and 24. For N = 24, IT = 0.9022 I0.
15-9. Using, λ/4 = (∆n) t,

λ 589.3 × 10−6 mm
t= = = 0.0162mm
4 ∆n 4 (1.5534 − 1.5443)
15-10. See Figure 15 24 that accompanies the statement of this problem in the text. Also refer to the figure below
for the labeling of the various angles:
D iagonal
′
θR
θR θ2
Ray
θ1
◦
45 θ4
θ3
Exit face
At the diagonal interface:

E p component from n k to n⊥: 1.4864 sin 45 = 1.6584 sin θR or θR = 39.329◦
′ ′
Es component from n⊥ to n k: 1.6584 sin 45 = 1.4864 sin θR or θR = 52.086◦
On exit:
Upper ray: θ1 = 45 − θR = 5.671◦; 1.6584 sin 5.671◦ = (1) sin θ2 or θ2 = 9.432◦
′
Lower ray: θ3 = θR − 45 = 7.086◦; 1.4864 sin 7.086◦ = (1) sin θ4 or θ4 = 10.566◦
Deviation: θ2 + θ4 = 9.432◦ + 10.566◦ = 19.997◦ ≈ 20◦
83
15-11. The arrangement is as shown below,
OA λ
∆ ϕ = π/2 or ∆ = = t ∆n
4
λ 600 × 10−6 mm
∆n = = = 0.005
t 4t 4 (0.03 mm)
15-12. See Figure 15 25 that accompanies the statement of the problem in the text.
(a) The incident angle is the polarizing angle,
n 1.33
tan θ p = 2 = ⇒ θ p = 53.12◦
n1 1
(b) The angle θR the refracted ray makes with the normal to the air/water interface is
sin θP

θR = sin−1 = 36.877◦
1.333
1.50

′
The polarizing angle for the water/glass interface is, θP = tan −1
= 48.37◦
1.333
If the glass surface was parallel to the water surface the angle of incidence on the glass would be
θR = 36.877◦. However, for complete polarization off the glass, θP′ must be 48.37◦. Thus the glass must be
tilted by 48.37◦ − 36.88◦ = 11.5◦ relative to the water surface.
15-13. (a) With polarizer alone rotating, there is no variation in intensity. This means there can be no linearly
polarized, or elliptically polarized, component. The light must be either unpolarized or circularly polar
ized, or a mixture of both. If pure circularly polarized, then insertion of a QWP should produce extinction
at some position. Since minima do not go to zero, the light is a mixture of circularly polarized and unpo
larized light. (See Meyer Arendt, for example, for a convenient diagrammatic analysis of such cases.)
(b) Here the rotating polarizer does produce a variation in intensity, so there must be linearly, or ellipti
cally, polarized light or a component of both. Since the intensity never goes to zero, it cannot be a case of
pure linearly polarized light. The use of the QWP now produces zero minima, so the light must be ellipti
cally polarized. Any linearly polarized component would not be extinguished under the condition
described.
15-14. (a) The angle of incidence at which maximum polarization occurs is,
2.42

θP = tan−1 = 56.2◦
1.62
(b) The angle of refraction into the diamond is
1.62 sin θP

θR = sin −1
= 33.8◦
2.42
15-15. The rotation of polarized light in an optically active medium is proportional to the inverse of the square of
the wavelength: β = ρ L d ∝ 1/λ2.
(a) For the given parameters, the concentration of the solution is
β 1.23
d= = g/cc = 0.05 g/cc
p L (20.5) (1.2)
(b) Given the stated wavelength dependence,
βred λ2
= violet ⇒ βviolet = βred λ2red /λ2violet = (15◦) (7002/4002) = 46◦
βviolet λ2red
84
πz
15-16. In general the thickness z is related to rotation β by, β = (nL − nR).
λ0
β λ0 (0.1745) (396.8 × 10−6mm)
(a) If β = 10◦ = 0.01745 rad, z = = = 0.200 m.
π (∆m) π (0.00011)
(b) Now, ρ = β for z = 1 mm. So,

πz π (1 mm)
ρ= ∆n = (0.00011) = 0.8709 rad = 49.9◦
λ0 396.8 × 10−6 mm
Also, by interoplation from Table 15 − 2, ρ ≈ 50◦.
λ λ 589.3 × 10−6 mm
15-17. (a) = t ∆n ⇒ t = = = 8.57 × 10−4 mm
4 4 ∆n 4 |1.4864 − 1.6584|
(b) For Zircon, using Table 15 1, n k = 1.968 and n⊥ = 1.923. It is given that t = 0.0182mm.
λ
To transmit, it must serve as a HWP so = t ∆n, or
z
λ = 2 t ∆n = 2 (0.00182 cm) (1.968 − 1.923) = 1.638 × 10−4 cm
This is in the infrared. In general
t ∆n 8.19 × 10−5 cm
(m + 1/2) λ = t ∆n ⇒ λ = =
m + 1/2 m + 1/2
8.19 × 10−5 cm
For m = 1, λ = = 546 nm (green). For m = 2, λ = 327.6 nm (ultraviolet). Thus in the visible,
1.5
green light is transmitted.
15-18. Let,
n2 n
tan θ p = and tan θp′ = 1
n1 n2
Then,
1
tan θ p = = cot θ p′
tan θ p′
So θ p = 90◦ − θ p′ . (See the triangle below.)
C tan α = cot β = A/B

β A
α α + β = 90◦
B
(b) See the figure below.
(1) tan θ p = n = n2/n1 − Brewster ’ s angle for light in.
θp
(2) sin θ p = n sin θR − Snell ’s law.
θR Then from (1): sin θ p = n cos θp.

Substituting 2): n sin θR = n cos θ p: θR and θ p must be complementary.
1 1 n
n = tan θ p = cot θR = and tan θR = = 1
tan θR n n2
θp
85
15-19. (a) ∆ = t ∆n. For t = λ, ∆/λ = ∆n = 1.5553 − 1.5462 = 0.0091
λ 546 × 10−6 mm
(b) = t ∆n or t = = 0.015 mm = 15µm
4 4 (.0091)
t ∆n (0.735 mm) (0.0091)
(c) m + 1/4) λ = t ∆n ⇒ m + 1/4 = = = 12.25 ⇒ m = 12
λ 546 × 10−6 mm
(d)
∆1 = − ∆2 ⇒ ∆total = t1 ∆n + t2 ( − ∆n) = λ/4
l 546 × 10−6 mm
t1 − t 2 = = = 0.015 mm = 15 µm
4 ∆n 4 (0.0091)
t1 t2
15-20. Given that λ = 500nm and t = 1/16 in = 0.0625 in = 0.158875cm,

(m + 1/2) λ (m + 1/2) (500 × 10−6 mm)
(m + 1/2) λ ∆n ⇒ ∆n = = = (m + 1/2) (3.1496 × 10−4)
t 1.5875mm
So for ∆m = 1 between adjacent bands, ∆n = 3.15 × 10−4.
λ0 ∆ϕ
15-21. From Eq. (15 3), d = . From Table 15 1, ∆n = 1.598 − 1.590 = 0.008.
2 π ∆n
(0.5893 µm) (π)
(a) Linear polarization: ∆ ϕ = π and d = = 36.8 µm
2 π (0.008)
π 36.8µ m
(b) Circular polarization: ∆ ϕ = and d = = 18.4 µm
2 2
15-22. From Table 15 2, ρ = 25.535 deg/mm at 546nm. Then

β = ρ L = + 25.535 deg/mm × 1.15 mm = + 29.365◦ (ccw)
Thus the HWP must be set at β/2 = + 14.68◦, since it converts linearly polarized light at + α to to lin
early polarized light at − α.
15-23. The polarizing angle is θ p = tan−1 n = tan−1 n2/n1 = tan−1 1.5 = 56.31◦.
(a) The reflectance is " √ #2

cos 56.31 − 1.52 − sin2 56.31
R= √ = 0.1479 ⇒ 14.79%
cos 56.31 + 1.52 − sin2 56.31
(b) The incident irradiance divides equally into I0/2 for each mode: Brewster’s angle is satisfied at each
interface. For the TE mode, 14.8% is reflected, so 85.2% is transmitted at each interface. The TM mode
is transmitted throughout at 100%. For N plates, there are 2 N interfaces, so for N = 10 we have
% TE transmitted = (0.852)2N I0/2 = 0.0203 I0 ⇒ 2.03%I0
(c) Since IT M = 50% of I0 ⇒ IT M = 0.50I0. So,

0.5 − 0.0203
P= = 0.922
0.5 + 0.0203
15-24. See Figure 15 26 that accompanies the statement of this problem in the text. The half wave plate rotates
the the polarization of the linearly polarized light so that it is polarized along a line that makes an angle of
2 θ with the x axis. Therefore the light polarization direction makes an angle of (90 − 2 θ) with the trans
mission axis of the analyzer. So, using Malus’ law, I = I0 cos2 (90 − 2 θ) ⇒ I = I0 sin2 (2 θ). his problem can
also be worked using matrices, in exactly the manner of problem 14 22, replacing the QWP matrix there by
the HWP matrix.
86
15-25. (a) Using Eq. (15 10) and Table 15 3,
πz π (3 × 10−3)
β= (nL − nR) = β = (6 × 10−5) = 0.742 rad = 42.5◦
λ0 762 × 10−9
(b) From Eq (15 3), with ∆ϕ = π,
π λ0 λ0
d= = =z
2 π|n⊥ − n k| 2 |n⊥ − n k|
So,
π λ0 π nL − nR π 6 × 10−5
β= (nL − nR) = = = 0.0105 rad = 0.60◦
λ0 2|n⊥ − n k| 2 |n k − n⊥| 2 8.94 × 10−3
87
Chapter 16 Holography
16-1. See Figure 16 9 that accompanies the statement of this problem in the text and the sketch below. Using
Eq. (7 14),
√ √
I = I1 + I2 + I1 I2 cos δ = I = I1 + I2 + 2 I1I2 2 cos2 (δ/2) − 1

with δ = 2 π ∆/λ. From the sketch, (s + ∆)2 = r 2 + s2 ⇒ s2 + 2 s ∆ + ∆2 = r 2 + s2 ⇒ ∆ (2s + ∆) = r 2

r2 π r2
So for ∆ ≪ s, ∆ ≈ and δ ≈ . Then,
2s λs
√ π r2

I1 I = I1 + I2 + 2 I1 I2 2 cos2
s
I2
y =r 2λs
 
!
I2 √ √ 2 πr
2
 4 I1 I2 cos 2 λ s 

O I = I1 + I2 − 2 I1 I2 + 
 
I2 B
I1 A αr 2
I = A + B cos2 (αr 2)
s
Plate
√
16-2. (a) As shown in problem 7 4, for (ER)2/(ES )2 = N , the visibility is V = 2 N /(N + 1)
√
(b) For N = 3, V = 2 3 /(3 + 1) = 0.866
16-3. If the storage could be done at the same density, the information stored holographically would be,
(1 Gb) (1mm/10−3mm) = 1000Gb
λn 633nm
16-4. d = = = 1823 nm = 1.82 µm
2 sinθ 2 sin 10◦
x λ/10 633 × 10−10 m

16-5. v = = −9 = = 63.3 m/s
t 10 s 10−9 s
2 2 2
Im ax AR + AS AR/AS + 1 8+1
16-6. = = = = 1.65
Imin AR − AS AR/AS − 1 8−1
V 1mm3 (10−3)3 m3
3 = (492 nm/1.30)3 = (492 × 10−9/1.30)3 m3 = 1.8 × 10 bits
10
16-7. Nb =
λn
16-8. (a) Refer to Figure 16 5 in the body of the text and note that 2 θ = 180◦. Then the spacing is given by
λn λ0 500 nm
d= = = = 156.25 nm
2 2 n (2) (1.6)
(b) For m λn = m λ0/n = 2 d sin 90◦ ⇒ . For m = 1, λ0 = 2 n d = 500 nm.
(c) In the film the angle with respect to the normal is sin ϕF = (1/1.6) sin (30◦) ⇒ ϕF = 18.21◦
The Bragg angle is then θ = 90◦ − ϕF = 71.79◦. The Bragg condition then gives, (for m = 1),
λ0
λn = = 2 d sin θ ⇒ λ0 = 2 n d sin θ = 2(1.6)(156.25 nm) sin (71.79◦) = 475nm
n
88
16-9. (a) See the sketch in Figure 16 7 for the orientation inside the film but note that the film being depicted
there has n = 1 and so the angle θ is the same inside and outside of the film. In air, the incident angle with
respect to the normal to the emulsion is ϕ0 = 30◦. The angle with respect to the normal in the film is then
sin ϕF = (1/n) sin ϕ0 = (1/1.6) sin (30◦) ⇒ ϕF = 18.21◦
so that the Bragg angle in the film is θ = 90◦ − ϕF = 71.79◦.
60◦ 60◦
18.21◦ 30◦
30◦
71.79◦
(b) The Bragg condition for m = 1 gives,
λn λ0 633 nm
d= = = = 208nm
2 sin θ 2 n sin θ 2(1.6) sin (71.79◦)
(c) Using the Bragg condition, the correct Bragg angle is,
λ0 450
sin θ = = = 0.6761 ⇒ θ = 42.54 ◦
2 n d 2 (1.6)(208)
The correct angle with respect to the normal within the film is then ϕF = 90◦ − 42.54◦ = 47.46◦. The angle
of incidence from air with respect to the normal is given by Snell’s law,
sin ϕ0 = 1.6 sin ϕF = 1.6 sin 47.46◦ = 1.18
No solution is possible for this wavelength and this film.
λ 430
16-10. Assuming a film of index of refraction unity. The original spacing is d0 = = = 215 nm. Upon
2 sin 90 2
shrinkage,
d = d0 − 0.15 d0 = 182.75 nm
Then, λ = 2 d sin 90◦ = 2 (182.75 nm) (1) = 365 nm. The blue components shift into the ultraviolet and
appear missing.
λR 633 λ 633
16-11. (a) m = = = 1.88 × (b) m = R = = 6330 ×
λuv 337 λx-ray 0.10
89
s i(ωt+δ)
16-12. (a) The plane and spherical waves are of the form ER = r eiωt and ES = e with EF = ER + ES .
R
Then,
2 s2 r s −iδ r s iδ
IF = |EF |2 = ER + ES2 + ER ES + ER ES = r 2 + 2 + e + e
R R R
2 2 2

s r s i(ωt−δ) r s i(ωt+δ)
EH ∝ IF ER = r2 + 2 ER + e + e
R R R
EH1 EH 3 EH 2
Here EH1 represents the amplitude modulated reference beam, EH3 represents the phase reversed subject
beam (real image), and EH2 represents the amplitude modulated subject beam (virtual image).
(b) The result δ = π y 2/λd was derived in problem 13 1 with d = s.
− ∆1
When the diverging spherical wave front (from P ) on
∆1 − ∆2 the left touches point C on the plate, phases of points
1′
1 ∆2 1 and 2 on the expanding wavefront for example
show lags related to the path difference:
2 2′ ∆1 > ∆2 > ∆0 = 0 .
P C P′ When these differences are reversed, i.e., promoted to

advances relative to C , then points such as 1 ′ and 2 ′,
shown along the direction of reinforcement, (toward P ′)
become appropriate. The points 0, 1 ′, 2 ′, ... are consistent
with the spherical wavefront converging toward the
Plate (real object) P ′.
90
Chapter 17 Optical Detectors and Displays
h c 1240 eV · nm 1240 eV · nm 1240 eV · nm

17-1. Egap = = ⇒λ= = = 1850 nm
λ λ Egap 0.67 eV
17-2. (a) 1/10=0.1.
(b) Assume that N electrons are generated per second. Then the responsivity is
N e/s e/s 1.6 × 10−19 C/s
R= = 0.1 = 0.1 = 0.0724 A/W
(10 N h c/λ)/s (h c/λ)/s (6.626 × 10−34 · 3 × 108/0.9 × 10−6) J/s
17-3. The responsivity is R = I/P ⇒ I = R P = (0.8 A/W) 10−7 W = 0.08 µA

6.626 × 10−34 · 3 × 108/1.5 × 10−6

Ephoton h c/λ
17-4. Pm in = = = W = 1.3 × 10−13 W
detection time tdet 10−6
17-5. (a) The number of photons Nph in a detection time tdet is,
NEP 4 × 10−11
Nph = = −34 = 300
(h c/λ)/tdet (6.626 × 10 · 3 × 108/1.5 × 10−6)/(10−6)
(b) Since the number of photons in the detection time is greater than one, the detector is not limited by
the intrinsic quantum fluctuations in the incident field.
17-6. The transmission through the liquid crystal cell is dependent on the voltage across the cell as indicated in
Figure 17 9. Modulating the voltage then modulates the transmitted irradiance.
17-7. See for example, Saleh, B. E. A., and M. C. Teich. Fundamentals of Photonics. New York: John Wiley and
Sons, 1991. Section 18 3. Phase modulation is most easily accomplished by using glass plates with parallel
scratches both scratched in the same direction. In the absence of a voltage across the cell the molecules
align in the direction of the scratches (θ = 0). With an applied electric field (directed from one glass plate
toward the other, the molecules tilt toward the direction of the field reaching an equilibrium tilt angle θe.
Since the liquid crystal is birefringent, the index of refraction seen by light polarized along the θ = 0 direc
tion is
1 cos2 θe sin2 θe
= +
n2(θe) n2e n2o
Modulating the voltage across the cell thus modulates the index of refraction by changing the tilt angle and
in turn the optical path difference and phase shift encountered in a pass through the cell are modulated.
17-8. The magnitude of the magnification is m = − s ′/s = − 10 cm/100 cm = 0.1. Thus a feature of object

dimension 1 mm will have an image dimension of 0.1mm. This the pixel spacing would have to be less than
0.1 mm. The number of pixels would have to exceed (20/0.1)2 = 40000.
17-9. The saturation power would be

Isat 10−5 A
Psat = = = = 0.1mW
R (0.1 A/W)
17-10. The light emerging from the calculator display is polarized. Rotating the polarizer thus changes the irradi
ance of the light reaching the eye. In one orientation the calculator display appears unchanged due to the
addition of the polarizer. When the polarizer is rotated 90◦ from this orientation, all of the light from the
display is blocked and the display appears dark.
91
Chapter 18 Matrix Methods in Paraxial Optics
18-1. See the sketch below. Using Eq. (18 1),

1 n − n ′ nL − n (nL − n) (nL − n ′) t
= L − −
f1 n R2 n R1 n nL R1R2
1 1.6 − 1 1.6 − 1.33 (1.6 − 1.33)(1.6 − 1) 5 cm
= − − ⇒ f1 = − 62.05 cm
f1 1.33( − 40 cm) (1.33) (40 cm) 1.33(1.6) (40 cm)( − 40 cm)
n′ 1
f2 = − f1 = − ( − 62.05 cm) = 46.66 cm
n 1.33
Further, using Eq. (18 3),
nL − n ′ 1.6 − 1
r= f1 t = ( − 62.05 cm) (5 cm) = 2.91 cm
nLR2 1.6 ( − 40 cm)
n −n 1.6 − 1.33
s=− L f2 t = (46.66 cm)(5 cm) = − 0.98 cm
nL R1 1.6( + 40 cm)
F1 n ′ = 1.33 nL n=1 F2
H2
f1
H1
f2
r s
18-2. (a) Using Eq. (18 1)

1 n − n ′ nL − n (nL − n) (nL − n ′) t
= L − −
1 1.53 − 1 1.53 − 1 (1.53 − 1)2 3 cm
= − − ⇒ f1 = 14.06 cm
f1 12.5 cm − 20 cm 1.53 (12.5cm)( − 20 cm)
n′
f2 = − f1 = − 1 ( − 14.06 cm) = − 14.06 cm
n
nL − n ′ 1.53 − 1
r= f1 t = (14.06 cm) (3 cm) = 1.17 cm
nLR2 1.53 (12.5 cm)
n −n 1.53 − 1
s=− L f2 t = ( − 14.06 cm)(3 cm) = − 0.73 cm
nL R1 1.53( − 20cm)
(b) With “object distance” 30 cm from V1 or s = 31.17 cm from H1,

1 1 1 1 1 1
+ = ⇒ + = ⇒ si = − 9.689 cm from H2
s0 si f 31.17 cm si − 14.06 cm
Since the lens center is 1.50 cm − 0.73 cm = 0.77 cm from H2, the image is − 9.689 cm + 0.77 cm = − 8.92 cm
from the lens center. Treating the lens as a thin lens,

1 1 1 1 1
= (n − 1) − = (1.53 − 1) + cm−1 ⇒ f = − 14.51 cm
f R1 R2 − 20 12.5
1 1 1 1 1 1
+ = = + = ⇒ si = − 9.780 cm from lens center
so si f 30 cm s ′ − 14.51 cm
The percent error in using the thin lens formula is
9.78 − 8.92
% error = = 9.6%
8.92
92
18-3. Using Eq. (18 1)
1 n − n ′ nL − n (nL − n) (nL − n ′) t
= L − −
1 1.5 − 1 1.5 − 1 (1.5 − 1)2 5 cm
= − − ⇒ f1 = 12.63 cm
f1 10 cm − 20 cm 1.5 (10 cm)( − 20 cm)
n′
f2 = − f1 = − 1 (12.63cm) = − 12.63 cm
n
nL − n ′ 1.5 − 1
r= f1 t = (12.63 cm) (5 cm) = 2.105 cm
nLR2 1.5 (10 cm)
n −n 1.5 − 1
s=− L f2 t = ( − 12.63 cm)(5 cm) = − 1.0525 cm
nL R1 1.5( − 20cm)
Then s0 = 8 cm + r = 10.105 cm and,
f1 f2 12.63 ( − 12.63 cm)
− + =1⇒− + = 1 ⇒ si = − 5.614 cm left of H2
so si 10.105 si
That is si = − 5.614 cm + s = − 6.67 cm left of V2. The magnification is,
n si (1)( − 5.614)
m=− =− = + 0.556
n ′ so (1) (10.105)
So, hi = m ho = 0.556 × 1in = 0.556 in.
18-4. Using Eq. (18 1)

1 n − n ′ nL − n (nL − n) (nL − n ′) t
= L − −
1 1.61 − 1.33 1.61 − 1 (1.61 − 1.33)(1.61 − 1) 2 cm
= − − ⇒ f1 = − 11.51 cm
f1 (1) ( − 10 cm) (1) (10 cm) (1) (1.61) (10 cm)( − 10 cm)
n′ 1.33
f2 = − f1 = − ( − 11.51cm) = 15.31 cm
n 1
nL − n ′ 1.61 − 1.33
r= f1 t = ( − 11.51 cm) (2 cm) = 0.4 cm
nLR2 1.61 ( − 10 cm)
n −n 1.5 − 1
s=− L f2 t = (15.31 cm)(2 cm) = − 1.16 cm
nL R1 1.61( 10cm)
Also, from Eq. 18 4,
n′ n − n′ 1.33 1.61 − 1.33

v= 1− + L t f1 = 1 − + (2) ( − 11.51 cm) = 4.20 cm
n nL R2 1 1.61 ( − 10 )
1.61 − 1

n nL − n 1
w= 1− ′ − t f2 = 1 − − 2 (15.31 cm) = 2.64 cm
n nL R1 1.33 1.61 (10)
Now, so = 60 cm + r = 60.4 cm. So,
f1 f2 − 11.51 15.31 cm
− + =1⇒− + = 1 ⇒ si = 18.9 cm from H2
so si 60.4 si
The magnification is,
n si (1)(18.9)
m=− =− = − 0.235
n ′ so (1.33) (60.4)
So, hi = m ho = ( − 0.235)(5 cm) = − 1.18 cm.
93
18-5. (a) Using Eq. (18 1)
1 n − n ′ nL − n (nL − n) (nL − n ′) t
= L − −
1 1.33 − 1 1.33 − 1 (1.33 − 1)2 20 cm
= − − ⇒ f1 = − 20.15 cm
f1 (1) ( − 10 cm) (1) (10 cm) (1) (1.33) (10 cm)( − 10 cm)
n′ 1
f2 = − f1 = − ( − 20.15 cm) = 20.15 cm
n 1
Using Eq. (18 3),
nL − n ′ 1.33 − 1
r= f1 t = ( − 20.15 cm) (20 cm) = 10 cm
nLR2 1.33( − 10 cm)
n −n 1.33 − 1
s=− L f2 t = − (20.15 cm)(20 cm) = − 10 cm
nL R1 1.33( 10cm)
Since n = n ′ = 1, v = r = 10 cm and w = s = − 10 cm.
(b) The object distance is so = 20 cm + 10 cm = 30 cm from H1. So,

f1 f2 − 20.15 20.15 cm
− + =1⇒− + = 1 ⇒ si = 61.38 cm from center
so si 30 si
s 61.38
The magnification is m = − i = − = − 2.05. The image is inverted, real, and approximately twice the
so 30
size of the object.
(c)
H1, N1
object
F2
F1
image
H2, N2
18-6. (a) The system matrix is

     
1 0 1 5 1 0 1 10/3
M = Rsph TRplane =    = 
− 1/10 3/2 0 1 0 2/3 − 1/10 2/3
y = A y0 + B α0 = (1)(1 cm) + (10/3) (0) = 1 cm

α = C y0 + D α0 = ( − 1/10) (1 cm) + (2/3) (0) = − 0.1 rad = − 5.73◦
With an additional translation,
    
1 x 1 10/3 1 − x/10 10/3 + 2 x/3
M ′ = T M =   = 
0 1 − 1/10 2/3 − 1/10 2/3
y = (1 − x/10) y0 + (10/3 + 2 x/3) α0

α = − 0.1 y0 + 2 α0/3
(c) The ray crosses the axis when y = 0. For y0 = 1, α0 = 0, y = 0 ⇒ 1 − x/10 = 0 ⇒ x = 10 cm
94
18-7. Use yf = A y0 + B α0 and α f = C y0 + D α0. From Figure 18 12b, a0 = 0, y0 = y0, y f = yf so,
y f = A y0, α f = C y0
Proceeding,y f = A y0 + B α0, α f = C y0 + D α0
yf A y0 A y y0 1
q≈ = = − ; f2 ≈ 0 = =−
− α f − C y0 C − α f − C y0 C
y − y f y0 − A y0 1 − A 1−A
−s≈ 0 = = ⇒s=
− αf − C y0 −C C
−A 1 1−A
Or, s = q − f2 = + ⇒s= . From Figure 18 12c, α0 = α f = α. Thus,
C C C
c y0
y f = A y0 + B α, α = C y0 + D α ⇒ α =
1−D
y A y0 + B α y y0 A − A D + B C A − Det(M ) A − n0/n f
−w≈ f = =A 0 +B =A +B= = =
α α α C y0/(1 − D) C V C
n0/n f − A
w≈
C
18-8. The lens matrix is

     
1 0 1 2 1 0 1/2 5/4
     
   = 
 1.6 − 1.3 1.6   1 − 1.6 1   
0 1 − 3/13 25/26
1.3 (1.5) 1.3 1.6 (1.5) 1.6
(1) (2)
p = D/C = − 25/6 = − 4.17 cm q = − A/C = 13/6 = 2.17 cm
n = 1.6
D − 1/1.3
n=1 n = 1.33
r= = − 0.83 cm s = (1 − A)/C = − 2.17 cm
2 cm C
n0/n f 1
f1 = = − 3.33 cm f2 = − = 4.33 cm
C C
R1 = 1.5 cm R2 = 1.5 cm
18-9. The lens matrix is,

     
1 0 1 5 1 0 1/2 5/4
    = 
1/10 1 0 1 − 1/10 1 − 1/20 3/2
f1 = 1/C = − 20 cm, f2 = − 1/C = 20 cm

q = − A/C = 10 cm, p = D/C = − 30 cm
r = (D − 1)/C = − 10 cm, s = (1 − A)/C = − 10 cm

     
1 0 1 3 1 0 4/5 2
     
1 =
     
 0.1 1.5   − 0.5 
0 1 − 3/70 11/14
1.4 (2) 1.4 1.5 (5) 1.5
f1 = (n0/n f )/C = (1/1.4)/C = − 16.67 cm, f2 = − 1/C = 25.33 cm

q = − A/C = 18.67 cm, p = D/C = − 18.33 cm
r = (D − n0/nf )/C = − 1.67 cm, s = (1 − A)/C = − 4.67 cm
95
18-11. (a), (b) The lens matrix is,
     
1 15 1 0 1 30 − 1/2 0
   = 
0 1 − 1/10 1 0 1 − 1/10 −2
f1 = 1/C = − 10 cm, f2 = − 1/C = 10 cm

q = − A/C = − 5 cm, p = D/C = 20 cm
r = (D − 1)/C = 30 cm, s = (1 − A)/C = − 15 cm
v = (D − 1)/C = 30 cm, w = (1 − A)/C = − 15 cm
The principal and nodal points fall at the lens center and the focal points fall 10 cm to either side of the
lens.
(c) This is case (c) of Figure 18 9. That B = 0 signifies that reference planes are object image, or conjugate,
planes and A represents the linear magnification.
Check by the thin lens Eq: If an object is at the input plane,
1 1 1
+ = ⇒ si = 15 cm
30 cm si 10 cm
Then m = − si/so = − 15/30 = − 1/2 = A.
18-12. (a) The lens matrix is,

     
1 15 1 8 1 30 1/3 16/3
   = 
0.5/( − 4) 3/2 0 1 ( − 0.5)/(4 × 1.5) 2/3 − 1/6 1/3
f1 = 1/C = − 6 in, f2 = − 1/C = 6 in

q = − A/C = 2 in, p = D/C = − 2 in
r = (D − 1)/C = 4 in, s = (1 − A)/C = − 4 in
(b) Parallel light focuses at F2, measured from the output plane (or right surface) by q = 2 in.
18-13. (a), (b) The lens matrix is,

     
1 0 1 1 1 0 16/15 2/3
     
   = 
   − 0.5   
0.5/( − 5) 3/2 0 1 2/3 − 1/150 14/15
(1.5) ( − 5)
f1 = 1/C = − 150 cm, f2 = − 1/C = 150 cm

q = − A/C = 160 cm, p = D/C = − 140 cm
r = (D − 1)/C = 10 cm, s = (1 − A)/C = 10 cm
96
18-14. (a), (b) The lens matrix is
       
 1 0  1 1/2 1 0 1 1 1 0 0.9764 0.96755
     
1 =
       
 0.62   − 0.1 1.52   − 0.52 
1.62 0 1 0 1 0.009182 1.0333
20 − 20 (1.62) 1.62 (20)(1.52) 1.52
f1 = 1/C = − 150 cm, f2 = − 1/C = 150 cm

q = − A/C = 160 cm, p = D/C = − 140 cm
r = (D − 1)/C = 10 cm, s = (1 − A)/C = 10 cm
(c) Consider each lens separately in air:
1.52 − 1

1 n2 − n1 1 1 1 1 1
= − ⇒ = − = 0.052 cm −1
1 .6 2 f n1 R1 R2 f1 1 20 cm − 20 cm
1.62 − 1

1 1 1
= − = − 0.062 cm−1
1 .5 2
f2 1 − 20 cm 20 cm
1 1 1
Equivalent focal length: = + = 0.052 cm−1 − 0.062 cm−1 ⇒ feq = − 100 cm
feq f1 f2
18-15. (a) Using the matrix calculated in the text,

     
1 s′ 2/3 2 1 s 2/3 − s ′/6 2/3 s + 2 − s s ′/6 + s ′
M =   = 
0 1 − 1/6 1 0 1 − 1/6 s/6 + 1
(b) Input and output planes correspond to object and image planes when B = 0. That is, when,
4 s + 12
2/3 s + 2 − s s ′/6 + s ′ = 0 ⇒ s ′ =
s−6
When B = 0, A = m = 2/3 − s ′/6.
(c) When A = 2/3 − s ′/6 = 0, the output plane at s ′ = 4 cm corresponds to the second focal plane. When
D = − s/6 + 1 = 0, the input plane at s = 6 cm corresponds to the first focal plane.
18-16. It is helpful to use a computer to perform the matrix multiplication in this problem... The sequence of
matrices has the form,
      
1 0 1 t3 1 0 1 d2 1 0 1 t2
      
      ×
 1 − n3   1 − n3 1   n2 − 1  
n3 0 1 0 1 n2 0 1
r6 n3 r 5 n3 r4
       
1 0 1 d1 1 0 1 t1 1 0 A B
       
× 1 =
      
 1 − n2 1   n1 − 1   1 − n1 
0 1 n1 0 1 C D
n2 r 3 n2 r2 n1r1 n1
Matrix multiplication and insertion of parameter values gives,
A = 0.93935, B = 22.2212, C = 0.009284, D = 0.8448
and
D A D−1 1−A
p= = − 90.99 mm, q = − = 101.18 mm, r = v = = 16.72 mm, s = w = = − 6.53 mm
C C C C
f1 = 1/C = − 107.71 mm, f2 = − 1/C = 107.71 mm
Since q is the distance of the focal point from the output plane, it is also the distance from the last surface
to the film plane.
97
      
1 0 1 t 1 0 1 d2 A B
      
    = 
 n − n′ nL   n − nL n    
L
0 1 0 1 C D
n ′ R2 n′ nL R1 nL
Performing the matrix multiplication gives,
nL − n ′ n t

n − nL nt n
A= t + 1, B = , D= ′ +
nL R1 nL n n ′ R 2 nL
nL − n ′ nL − n (nL − n) (nL − n ′) t
C= − ′ −
n ′ R2 n R1 n ′ nL R1 R2
n 1 1 n′
18-18. From Table 18 2, f1 = ′
⇒ = C. Substituting C from problem 18 17,
n C f1 n
n ′ nL − n ′ nL − n (nL − n) (nL − n ′) t

1
= − ′ −
f1 n n ′ R2 n R1 n ′ nL R1 R2
1 n − n ′ nL − n (nL − n) (nL − n ′) t
= L − −
f1 n R2 n R1 n nL R1 R2
18-19. From Table 18 2 and using the results of problem 18 17,

h i
n nL − n ′ nt n
D − n/n ′
n′
+ n ′ R2 nL
− n′
r= =
C C
n 1
but, f1 = . So,
n′ C
nL − n ′ n′ nL − n ′

nt
r= f1 = f1 t
n ′ R2 nL n nL R2
Similarly,
h i
n − nL
1−A 1− nL R1
t+1 n − nL n −n
s= = = t ( − f2) = − L f2 t
C C nL R 1 nLR1
and,

n nL − n ′ nt
D −1 n′
+ n ′ R2 nL
−1
n

nL − n ′ n t
′
n n ′ nL − n ′

v= = = + − 1 f 1 = 1 − + t f1
C C n′ n ′ R2 nL n n nL R2
(n/n ′) − A

n n − nL n nL − n
w= = ′− t − 1 ( − f2) = 1 − ′ − t f2
C n nL R1 n nL R1
98
18-20. A typical program in “Basic” which can be easily adapted to other languages:
10 REM: MERIDIONAL RAY TRACING. AFTER THE LAST SURFACE IS CALCULATED, ENTER ANY
20 REM: NEGATIVE NUMBER AS DISTANCE TO NEXT SURFACE TO TERMINATE
30 INPUT ′′ LEFT AND RIGHT REFERENCE INDICES: ′′,N1,N2
40 INPUT ′′RAY ANGLE DEGREE AND HEIGHT: ′′,A,H
50 INPUT ′′DISTANCE FROM OBJECT POINT (IF INFINITY ENTER ANYTHING): ′′,D
60 REM: FOR RAY ENTERING PARALLEL TO AXIS, D IS NOT USED IN THE CALCULATION
70 P=3.1415926#:A=P*(A/180):REM ANGLE IS NOW IN RADIANS
80 IF A=0 THEN GOTO 110
90 S=D H/TAN(A):Q= S*SIN(A)
100 GOTO 140
110 Q=H
120 REM: FOR PLANE SURFACE, ENTER R>99990.
130 REM: THE ACTUAL VALUE IS NOT USED IN THE CALCULATION.
140 INPUT ′′INPUT RAD OF CURV (99999 IF PLANE): ′′,R
150 IF R>99990! THEN GOTO 210
160 B=Q/R+SIN(A): T=ATN(B/SQR(1 B*B)):REM TAKING ARCSIN(B)
170 C=N1*SIN(T)/N2:T1=ATN(C/SQR(1 C*C)):REM TAKING ARCSIN(C)
180 A1=T1 T+A:A2=(A1/P)*180:REMA2 CONVERTS A1 TO DEGREES AGAIN
190 Q1=(SIN(T1) SIN(A1))*R:S1= Q1/SIN(A1)
200 GOTO 240
210 X=N1*SIN(A)/N2:A1=ATN(X/SQR(1 X*X)):REM TAKING ARCSIN(X)
220 Q1=Q*COS(A1)/COS(A)
230 S1= Q1/SIN(A1):A2=(A1/P)*180:REM A2 IS A1 IN DEGREES
240 PRINT ′′RAY INTERSECTS AXIS AT ′′;S1; ′′FROM SURFACE. ′′
250 PRINT ′′RAY ANGLE IN RADIANS IS ′′;A1
260 PRINT ′′RAY ANGLE IN DEGREES IS ′′;A2
270 PRINT ′′THE Q PARAMETER IS ′′;Q1
280 A=A1:N1=N2
290 INPUT ′′DISTANCE TO NEXT SURFACE: ′′ ,D1
300 IF D1<0 THEN 360
310 Q=Q1+D1*SIN(A1)
320 INPUT ′′NEW REFRACTIVE INDEX: ′′,N2
330 INPUT ′′NEW RADIUS OF CURV (99999 IF PLANE): ′′,R
340 IF R>99990! THEN GOTO 210
350 GOTO 360
360 END
18-21. Use the computer program (problem 18 20) or Table 18 3 in a two step process for each ray, yielding
α = 0◦: s ′ = 3.180 cm and α ′ = − 23.51◦
α = − 20◦: s ′ = 16.104 cm and α ′ = 6.081◦
18-22. Use the computer program (problem 18 20) or Table 18 3 in a 3 step process, giving,
s ′ = − 49.525 cm and α ′ = 3.371◦
18-23. Use the computer program (problem 18 20) or Table 18 3 in a 6 step process, to get,
h = 1: s ′ = 98.20 mm and α ′ = − 0.567◦
h = 5: s ′ = 102.45 mm and α ′ = − 2.723◦
99
Chapter 19 Optics of the Eye
19-1. (a) The conversion between the radiant flux and the luminous flux is
Φv = K(λ) Φe = 685 (λ) Φe
Estimating from Figure 19 2,
Φv,He-C d 685 (0.02) (0.050)
= = 1.25
Φv,He-Ne 685 (0.2) (0.004)
(b) For equal brightnesses,
Φv,Ar = Φv,He-Ne ⇒ 685 V (λAr ) Φe,Ar = 685 V (λHe-Ne Φe,He-N e

(0.2) Φe = (0.95) (0.5 mW)

Φe,Ar = 2.4 mW
19-2. (a) From Table 19 1,

Iv dΦv/dω Φv/4 π
= = ⇒ Iv = Ev r 2 = (100 lm/m2) (3 m)2 = 900cd
Ev dΦv /dA Φv/(4 π r 2)
(b) See Eq. (1 13), Iv (θ) = Iv (0) cos θ.
Source Using the general result in (a), at point Q In the diagram,

√ √
r= 10
3m I (0) cos θ (900 lm) (3/ 10 )
θ Ev = v 2 = √ 2 = 85.4 lx
r
P Q Flo or 10 m
1m
19-3. Use the relation between illuminance and luminous intensity given in the solution to problem 19 2a:
Iv(0) cos (0) Iv(0) 3/5

= 1.351 × 10−3/ft2 Iv(0)

Ev(P ) = +
S1
√
S2 (30 ft)2 (50 ft)2
1300 ′  √ 
50 ′ I
 v (0)( 30/ 1300
 = 1.280 × 10−3/ft2 Iv (0

Ev(Q) = 2

30 ′ √ 2
1300 ft
Ev (P )
P
20 ′ Q = 1.055
Ev(Q)
19-4. The arrangement is sketched (not to scale) below,
The indicate angle and solid angle satisfy the relations
cos θ0 = cos(sin−1(5/50) = 0.9950, Ω0 = 2 π (1 − cos θ0 = 0.01 π sr

50 cm
θ0 5 cm
r
A F Ω0 Φv,inc = Ω0 Iv = (0.01π) (100) = π lm, Φv,ref = 0.8 Φv,inc = 0.8 π lm

Φv,ref 0.8 π lm
Ev = = = 320 lx
A π (0.05 m)2
100
19-5. (a) The solid angle intercepted is Ω = 2 π (1 − cos θ) = 2 π (1 − cos(0.25◦)) = 5.98 × 10−5 sr. Here θ is the half
angle intercepted by the sun viewed from the earth. The luminance of the sun is about,
Ev 105 lm/m2
Lv = = = 1.67 × 109 cd/m2
Ω 5.98 × 10−5 sr
θ
E S
lm

(b) Ev = Lv Ω = L × 2 π sr = 2 π L lm/m2.
m2 sr
19-6. The arrangement is sketched below.

r
A1
r1 2 θ2
dA1 θ1 45 ◦
D=1m ∆A2 = 1 cm 2
The luminous flux incident on the small surface is

L cos θ1 cos θ2 cos θ1 cos θ2
Z Z Z
∆Φ12 = 2 dA1 dA2 = L ∆A2 2 dA1
A1 A2 r 12 A1 r12
Consider first illumination of ∆A2 by an annular ring of incremental area dA1 = 2 π r dr, as shown. Then,
√
r12 = r 2 + D 2, cos θ1 = D/ r 2 + D 2
√
2 D r
cos θ2 = cos(45 − θ1) = cos(45 ) cos θ1 + sin (45 ) sin θ1 =
◦ ◦ ◦ √ +√
2 r2 + D2 r2 + D2
Adding the contributions from the annular rings that make up the disc,
Z R
cos θ1 cos(45◦ − θ1)
∆Φ12 = L ∆A2 2 π r dr
0 r2 + D2
Z R
√
D 2 D r 2πr
∆Φ12 = L ∆A2 √ √ +√ dr
0 r +D 2
2 2 2
r +D 2 2
r +D 2 r + D2
2
( Z )
√ R R
r 2 dr
Z
r dr
∆Φ12 = 2 π L D ∆A2 D +
0 (r 2 + D2)2 0 (r + D2)2
2
( R )
√ −1 −r 1 R
−1 r
∆Φ12 = 2 π L D ∆A2 D + + tan
2 (r 2 + D 2) 0 2 (r2 + D 2) 2 D D 0
√

1 −1 R 1 D+R
∆Φ12 = 2 π L D ∆A2 tan + −
2D D 2 D 2(R2 + D2)
With L = 105 lm/sr m2, D = 1 m, R = 0.2 m, dA2 = 10−4 m2,
∆Φ12 = 0.97 lm
n2 − n1
19-7. The power for a refracting surface is defined as P = . So,
R
nA.H − 1 1.33 − 1
P= = = 41.6 m
R 8 × 10−3 m
This value is in agreement with Table 19 2.
101
19-8. From Table 19 2: nlens = 1.45, R1 = 10 mm, and R2 = − 6 mm.
(a) Using the lensmaker’s formula,
1.45 − 1

1 nlens − 1 1 1 1 1 1 1
= − = − ⇒ f = 8.33 mm = P = = = 120 D
f 1 R1 R2 1 10 mm − 6 mm fin m 0.00833 m
(b) In its environment,
1.45 − 1 1

1 nlens − nA H 1 1 1 1
= − = − mm−1 ⇒ f = 41.7 mm = P = = 24.0 D
f nA H R1 R2 1 10 − 6 fin m
(c) For a thick lens consider the two successive refractions and the translation across the lens,
For f1: s2′ = ∞, s1 = f1 (measured from the first surface). Then,

R1 = 10 m m
R2 = − 6 m m nlens nAH nA H − nlens 1 1.33 − 1.45
+ = ⇒ = = s2 = 66.7 mm
nA H = 1.33 nA H = 1.33 s2 ∞ R2 s2 (1.33) ( − 6 mm)
F2
s1′ = ( − 66.7 − 3.6) mm = − 63.1 mm.
F1 f2 f1
nL
nAH nlens nlens − nAH
nl e n s = 1.45 + ′ = ⇒ s1 = f1 = 38 mm
3.6 m m s1 s1 R1
Proceeding in the same way for f2:
s1 = ∞, s2′ = f2 (measured from the second surface)
nAH nlens nlens − nAH

+ ′ = ⇒ s1′ = 120.8 mm
∞ s1 R1
s2 = − (120.8 − 3.6) mm = − 117.2 mm
nlens nAH nlens − nAH

+ = ⇒ s2 = f2 = 41.1 mm
s2 s2 R2
Note that these focal lengths are measured from the lens surfaces. The more formal matrix method leads to
focal lengths measured from the principal planes . Use the techniques of Chapter 18, The lens matrix is,
     
1 0 1 3.6 1 0 0.9702 3.3021
     
1.33 =
     
 1.45 − 1.33 1.45   1.33 − 1.45 
0 1 − 0.02361 0.9503
(1.33) ( − 6) 1.33 (1.45) (10) 1.45
Surface R2 Translation Surface R2
q = − A/C = 41.09 mm, p = D/C = − 40.25 mm

r = (D − 1)/C = 2.105 mm, s = (1 − A)/C = − 1.262 mm
f1 = 1/C = − 42.355 mm f2 = − 1/C = 42.355 mm
102
19-9. Refer to Figure 19 3 and Table 19 2.
(a) For an unaccommodated eye the object is at infinity. Then at the cornea
n1 n2 n2 − n1 n R (1.33) (8 mm)
+ ′= ⇒ s′ = 2 = = 32 mm (right of cornea)
s s R n2 − n1 1.33 − 1
At the front surface of the lens which is 3.6 mm from the cornea: s = (3.6 − 32) mm = − 28.4 mm.
nAH nlens nlens − nAH 1.33 1.45 1.45 − 1.33
+ ′ = = + ′ = ⇒ s ′ = 24.73 mm (right of 1st lens surface)
s s R1 − 28.4 mm s 10 mm
At the rear surface of the lens, s = (3.6 − 24.73 )mm = − 21.13 cm:
nlens nAH nAH − nlens 1.45 1.33 1.33 − 1.45
+ ′ = ⇒ + ′ = ⇒ s ′ = 15.13 mm (right of 2nd lens surface)
s s R2 − 21.13 mm s − 6 mm
Thus an object at infinity forms an image at 15.13 mm + 7.2 mm = 22.33 mm from the cornea the back focal
plane of the eye. This value is in agreement with Table 19 3.
(b) Repeating for the parameters associated with an eye accommodated for an object that is 25 cm from
the cornea:
At the cornea:,
1 1.33 1.33 − 1
+ ′ = ⇒ s ′ = 35.43 mm (right of cornea)
250 mm s 8 mm
At the front surface of the lens, s = (3.2 − 35.43) mm = − 32.33 mm:
1.33 1.45 1.45 − 1.33
+ ′ = ⇒ s ′ = 23.83 mm (right of front lens surface)
− 32.53 mm s 6 mm
At the rear lens surface, s = (4 − 23.83) mm = − − 19.83 mm:
1.45 1.33 1.33 − 1.45
+ ′ = ⇒ s ′ = 14.39 mm (right of back lens surface)
− 19.83 s − 6 mm
So, for an eye accommodated for the normal near point, the image is formed at
(14.39 + 7.2) mm = 21.59 mm form the cornea.
19-10. (a) The matrix for propagation from the cornea to the back lens surface is,
     
1 0 1 3.6 1 0 1 3.6 1 0
     
M = 1.45 − 1.33 1.45 
   
 1.33 − 1.45 1.33  
  
  1 − 1.33 1


0 1 0 1
(1.33) ( − 6) 1.33 (1.45) (10) 1.45 (1.33) (8) 1.33
R ear lens surface Lens Trans. Front lens surface Trans. to cornea Cornea surface
 
0.75846 5.1050
M = 
− 0.05012 0.65180
q = − A/C = 15.133 mm, p = D/C = − 13.005 mm

r = (D − 1)/C = − 1.963 mm, s = (1 − A)/C = − 4.820 mm
f1 = (n0/n f )/C = (1/1.33)/C = − 15.00 mm f2 = − 1/C = 19.95 mm
The first principal plane (H in Figure 19 3) is a distance |r| = 1.96 mm to the right of the front corneal sur
face. The second principal plane (H ′ in Figure 19 3) is a distance |s| = 4.82 mm to the left of the back lens
surface which places it a distance of 7.2 mm − 4.82 mm = 2.38 mm to the right of the back lens surface.
These positions are in agreement with the values in Table 19 3. The front focal point is a distance of 15.0
mm to the left of the H which places it a distance of 13.0 mm from the front corneal surface. The back
focal plane is a distance of 19.95 mm to the right of H ′ which places to the right of the corneal surface by a
distance of 19.95 mm + 2.38 mm = 22.3 mm. These positions also agree with the values given in Table 19 3.
103
19-11. In a Snellen chart letters in the 20/20 row should subtend an angle of θl = 5 ′ = 5/60◦ = 0.001454 rad and
details should subtend an angle θd = 1 ′ = 0.0002909 rad. Heights on the screen a distance of D = 5 ft from
the eye are related to these angles by the relation y = D θ. So
20/20 row: yl = D θl = (60 in) (0.001454 rad) = 0.0872 in, yd = yl/5 = 0.0174 in.
20/15 row: yl = (15/20)yl,20/20 = 0.00654 in, yd = yl/5 = 0.00131 in.
19-12. Objects 25 cm from the eye should form images 125 cm from the eye. Since the lens is 25 cm from the eye
this implies object and image distances of s = 23.5 cm and s ′ = − 123.5 cm.
(a) The needed power is then
1 1 1 1 1
P= = + = − = 3.45/m = 3.45 D
f s s ′ 0.235 m 1.235 m
(b) Assuming that the eye alone has a far point at infinity, the far point with the lenses is the object posi
tion at which the corrective lenses form an image at − ∞. The object distance sfar from the glasses is
found as,
1 1 1
P= = + ⇒ sfar = f = 1/P = (1/3.45) m = 0.290 m = 29.0 cm
f sfar − ∞
This corresponds to a distance of 30.5 cm from the eye. The range of clear vision is expanded if the correc
tion is for a distance further than the normal near point.
19-13. (a) The lens should place the image of an object at infinity at the eye’s far point. So,
1 1 1 1 1 1
P= = + = + = =−2D
f s s ′ ∞ s ′ − 0.5 m
(b) The new near point is the object position that places an image 15 cm from the eye:
1 1 1 1
P = − 2/m = + = − ⇒ s = 0.214 m = 21.4 cm
s s ′ s 0.15 m
(c) Adjusting the procedure for a lens 2 cm from the eye:
1 1 1 1 1 1
P= = + = + = = − 2.083 D
f s s ′ ∞ s ′ − 0.48 m
1 1 1 1
P = − 2.083/m = + ′ = − ⇒ s = 0.178 m = 17.8 cm
s s s 0.13 m
The needed power is − 2.083 D and the near point is 17.8 cm from the glasses or 19.8 cm from the eye.
19-14. (a) − 1.50 D to correct for myopia; − 1.50 D to correct for astigmatism, cylinder axis horizontal
(b) − 2.00 D to correct for simple myopia
(c) + 2.00 D to correct for simple hyperopia
(d) + 2.00 D to correct for hyperopia; − 1.50 D for astigmatism, cylinder axis horizontal
104
19-15. The unaided far point is the image position for objects at infinity. The unaided near point is the image
position for objects at the corrected near point.
1 1 1 1 1 1 1 1
(a) Right eye far point: P = = − 7 D = + ′ = + ⇒ s′ = f = = = − m = − 14.3 cm
f s s ∞ s′ P −7D 7
The unaided far point is 14.3 cm from the eye.
1 1 1 1 1
Right eye near point: P = =−7D= + ′ = + ⇒ s ′ = − 0.0732 m = − 7.32 cm
f s s 0.15 m s ′
The unaided near point is 7.32 cm from the eye.
1 1 1 1 1 1 1 1
Left eye far point: P = = − 5D = + ′ = + ⇒ s′ = f = = = − m = − 20 cm
f s s ∞ s′ P −5D 5
The unaided far point is 20 cm from the eye.
1 1 1 1 1
Left eye near point: P = =−5D= + ′ = + ⇒ s ′ = − 0.0857 m = − 8.57 cm
f s s 0.15 m s ′
The unaided near point is 8.57 cm from the eye.
(b) The corrected far point is the object position that places an image at the unaided far point. The cor
rected near point point is the object position that places an image at the unaided near point. With the
wrong lens, then:
1 1 1 1 1
Right eye far point: P = =−5D= + ′ = + ⇒ s = 0.502 m = 50.2 cm
f s s s − 0.143 m
The far point with the wrong lens is 50.2 cm.
1 1 1 1 1
Right eye near point: P = =−5D= + ′ = + ⇒ s = 0.115 m = 11.5 cm
f s s s − 0.0732 m
The near point with the wrong lens is 11.5 cm.
1 1 1 1 1
Left eye far point: P = =−7D= + ′ = + ⇒ s = − 0.5 m
f s s s − 0.20 m
This answer is nonsense as the object position can not be negative for a single lens. This means that the
eye must strain to see object at infinity. The relaxed eye will not see anything clearly. So the far point
with the wrong lens is infinity but the eye must strain to see the far away objects.
1 1 1 1 1
Left eye near point: P = =−7D= + ′ = + ⇒ s = 0.214m = 21.4 cm
f s s s − 0.0857 m
The near point with the wrong lens is 21.4 cm.
19-16. Lets choose the lens to eye distance to be 1.5 cm. The lens for far vision should place images of far away
object a distance of (15 − 1.5) cm from the lens:
1 1 1 1
Pfar = + = + = − 7.41 D
s s ′ ∞ − 0.13.5 m
With this lens alone the corrected near point would be the object distance leading to images 11.5 cm from
the lens:
1 1 1
= P − ′ = − 7.41 D − ⇒ s = 0.778 m
s s − 0.115 m
Bifocals are needed. Let’s say that the comfortable reading distance is to be 40 cm from the eye. Then the
lens for near vision should have a power:
1 1 1 1
Pnear = + = + = − 6.10 D
s s ′ 0.385 m − 0.115 m
105
Chapter 20 Aberration Theory
20-1. Using Eq. (20 6),

a(Q) = (n1 l + n2 l ′) − (n1 s + n2 s ′) (1)
Eqs. (20 16) and (20 17) are,
" #
h2 (R + s) h4(R + s) h4 R + s)2
l=s 1+ + − (2)
2 R s2 8 R 3 s2 8 R 2 s4
" #
2 ′ 4 ′ 4 ′ 2
h (R − s ) h (R − s ) h R − s )
l′ = s′ 1 + + − (3)
2 R s ′2 8 R3 s ′2
′
8 R2 s 4
Using these in (1) and doing some obvious simplification gives,
" # " #
h2 (R + s) h4(R + s) h4 R + s)2 h 2
(R − s ′
) h 4
(R − s ′
) h 4
R − s ′ 2
)
a(Q) = n1 s + − + n2 s ′ + − (4)
2 R s2 8 R 3 s2 8 R 2 s4 2 R s ′2 8 R3 s ′2
′
8 R2 s 4
Consider the terms proportional to h2:

h2 n1 (R + s) n2(R − s ′) h2 n1 n2 n1 − n2

+ = + ′+ =0
2 Rs R s′ 2 s s R
Now the h4 terms:
h4

n1 n1 n1 n1 n1 n2 n2 n2 n2 n2
+ − − 2 − + − − + 2 − =
8 R 2 s R 3 s3 R s2 R2s R2s ′ R3 s ′3 R s ′2 R2)
s′
( 2 2
h4

n1 1 1 n1 1 1 1 n1 n2 n1 n2
− + − ′ − + + ′+ −
8 s s R s s′ R R2 s s R R
( )
2 2
h4 n1 1 1

n 1 1
a(Q) − + + 1′ −
8 s s R s s′ R
20-2. (a) If
n1 n2 n2 − n1 1 1 1
+ ′= and ′ = +
s s R s s R
then,
n1 n2 1 1 n n n n
+ ′ = (n2 − n1 − = 2′ − 2 − 1′ + 1
s s s′ s s s s s
n2 n1 ′
= − ′ ⇒ s = − (n1/n2) s
s s
(b) In a(Q) the h2 term is zero due to Gaussian imaging. For the remainder we must examine
( 2 2)
h4 n1 1 1

n2 1 1
a(Q) = − + + ′ −
8 s s R s s′ R
Substitute using 1/s ′ = 1/s + 1/R, to give,
( 2 2)
h4 n1 1 h4 n1 n1 s + n2 s ′

1 1 n2 1 1 1 n2
a(Q) = − − + + ′ + − =− + = =0
8 s s′ R R s s R R 8 s s ′2 s ′s2 s ′2s2
(c) For s ′ = R, s must be − R. These values make 1/s + 1/R and 1/s ′ − 1/R vanish independently so
a(Q) = 0. Also a(Q) = 0 when h = 0, for rays incident at the vertex (s = s ′ = 0).
1 1 1
(d) For R = 8 cm, = + and s ′ = − (1.36/1.70) s = − 0.8 s
s ′ s 8 cm
Solving these simultaneously gives, s = − 18 cm and s ′ = 14.4 cm. Thus aplanatic images occur at
s ′ = 0, 8 cm, and 14.4 cm.
106
20-3. The paraxial image point is found from
n1 n2 n2 − n1
+ ′=
s s R
since s = ∞,
1 1 − 1.5
0+ = ⇒ s ′ = 80 cm
s ′ − 40 mm
Then the ray aberration is becomes, (with s = ∞,)
( 2 2) 2 2
h4 n1 1 1 h4 n2′ 1 254 1 1

n2 1 1 1 1
a(Q) = − + + ′ − = − − = − − mm
8 s s R s s′ R 8 s′ s′ R 8 80 80 − 40
a(Q) = − 0.858 mm
Proceeding,
2 2
h3 n2 1 253 1 1

da 1 1
=− − = − − = − 0.1373
dh 2 s′ s′ R 2 80 80 − 40
s ′ da 80 mm
by = = ( − 0.1373) = − 10.98 mm
n2 dh 1
s by 80
′
bz = = ( − 10.98 mm) = − 35.14 mm
y 25
20-4. Refer to the figure below:
a(Q) = PQI − POI = ℓ + ℓ − (s + s ′).

In triangle PQC:
Q ℓ2 = R2 + (s − R)2 + 2 R (s − R) + 2 R (s − R) cos φ
In triangle CQI:
ℓ
ℓ′ h ℓ ′2 = R2 + (R − s ′)2 − 2 R (R − s ′) cos φ
R
φ p 1/2 h h4
Now cos ϕ = 1 − sin2 ϕ = 1 − (h/R)2
O

≈1− 2
−
P C I F 2R 8 R4
s′
R This expression can be used in the expressions for ℓ and ℓ ′ and
the resulting expressions expanded. For this task I will ask for
s
help from Maxima:
(%i1) (F:1-h^2/2/R^2-h^4/8/R^4,display(F))
h2 h4
F =− − +1
2 R2 8 R4
(%i2) (L:sqrt(R^2+(s-R)^2+2*R*(s-R)*F),display(L))
s
h2 h4

L = R2 + 2 − 2
− 4
+ 1 (s − R) R + (s − R)2
2R 8R
(%i3) (Lp:sqrt(R^2+(R-sp)^2-2*R*(R-sp)*F),display(Lp))
s
h2 h4

Lp = (R − sp)2 + R2 − 2 − 2
− 4
+ 1 R (R − sp)
2R 8R
(%i4) (Lapp:taylor(L,h,0,4),’L Lapp)
(R − s) h2 R3 − 2 s R2 + s3 h4

(%o4) L = s + − + ...
2sR 8 s3 R3
(%i5) (Lpapp:taylor(Lp,h,0,4),’Lp Lpapp)
(R − sp) h2 R3 − 2 sp R2 + sp3 h4

(%o5) Lp = sp + − + ...
2 sp R 8 sp3 R3
The last two expressions are ℓ and ℓ ′ to 4th order in h

continued ...
107
20-4. continued...
Using the expressions from Maxima in the expression for a(Q) gives,

3 ′ 2 ′3
(R − s) h 2 R 3
− 2 s R 2
+ s 3
4
h ′
(R − s ) h 2 R − 2 s R + s h4
a(Q) = ℓ + ℓ ′ − s − s ′ = − + −
2sR 8 s3 R 3 2 s ′R 8 s ′3 R3
h2 1 1 h4 1

2 1 1 1 1 1
a(Q) = + ′− − + − 2 − 2 + +
2 s s R 8 s3 s ′3 s2 R s ′2R R3 R3
Gaussian imaging gives,
1 1 2
+ − =0
s s′ R
So, the term proportional to h2 vanishes. This relation can be used to eliminate s ′ in the h4 term:
" 3 2 #
h4 1

2 1 2 1 2 2 1 2
a(Q) = − + − − − − + 3
8 s3 R s R s2 R R s R
2
h4 1 1

a(Q) = − −
4R s R
20-5. Using the result from problem 20 4,
2
h4 1 1

a=− −
4R s R
h4 (25)4
With s → ∞, a = − = − cm = − 0.153 cm = 0.0153 mm
4 R3 4 (400)3
da h3 253
Further, =− 3 =− = − 2.441 × 10−4
dh s=∞ R 4003
Then
s′ da
= (200 cm) − 2.441 × 10−4 = − 0.0488 cm = − 0.488 mm

by = =f
n2 dy
s′ fb 200 ( − 0.0488)
bz = b y = y = cm = − 0.391 cm = − 3.91 mm
y h 25
20-6. As in problem 20 5, with s = ∞, a = − h4/ 4R3 . Also, h = f /(2 × 3.75) = 40 cm

h4 (40)4
(a) a = − 3
= cm = − 0.00296 cm = − 0.0296 mm
4R − 4 (600)3
(b) n ∆t = |a| ⇒ ∆t = |a|/n = (0.0296 mm)/1.4 = 0.021 mm
20-7. Given: bz = 1 cm, 4 D (f = 25 cm), 6 cm diameter lens (h = 3 cm). For s = 50 cm, s ′ = s f /(s − f ) = 50 cm.
da hb (3) (1)
(a) by = − s ′ = − ′z = − cm = − 0.6 mm (b) dblur = 2 |bx | = 2 |b y | = 2 (0.6 mm) = 1.2 mm
dy s 50
r2 + r1 − 10 + 10
20-8. The Coddington shape factor is σ = = = 0 and the focal length is,
r2 − r1 − 10 − 10

1 1 1
= (1.5 − 1) − ⇒ f = 10 cm
f 10 cm − 10 cm
For parallel rays, s p′ = f = 10 cm. Then, from Eq. (20 28)
s ′ − s s/s ′ − 1
p= ′+s
= = − 1 (as s → ∞)
s s/s ′ + 1
1.53

1 1 1 1
′ − = 0 + 0 + [3(1.5) + 2] [1.5 − 1] ( − 1)2 + ⇒ sh′ = 9.836 cm
sh 10 cm (8)10 (1.5) (0.5 ) cm
3 1.5 − 1
bz = s p′ − sh′ = (10 − 9.836) cm = 0.164 cm; by = h/s ′) bz = 0.0164 cm
108
20-9. The focal length is given by the lensmaker’s formula

1 1 1
= (1.6 − 1) − ⇒ f = 20 cm = s p′
f 36 cm − 18 cm
s′ − s r + r1 − 18 + 36
Also, since s → ∞, p = = − 1 and σ = 2 = = − 1/3. So, using Eq. (20 28)
′
s +s r −r − 18 − 36
( 2 1 )
2
h2 3.6 1.63

1 1 1 1
− = − + 4 (2.6) − ( − 1) + 6.8 (0.6) ( − 1) +
2
sh′ 20 cm 8 (20)3 (1.6) (0.6) cm 6 3 3 0.6
1
sh′ =
0.05/cm + 0.00024479 h2/cm3
1
bz = 20 cm − sh′ = 20 cm −
0.05/cm + 0.00024479 h2/cm3
Plotting gives,
3 h (cm) bz (mm)
1 0.974
2.5
2 3.84
2 3 8.44
bz (cm)
1.5 4 14.53
1
5 21.81
0.5
0
0 1 2 3 4 5 6
h (cm)
r2 + r1 − 10 + 10
20-10. The Coddington shape factor is σ = = = 0 and the focal length is,
r2 − r1 − 10 − 10

1 1 1
= (1.5 − 1) − ⇒ f = 15 cm
f 15 cm − 15 cm
s ′ − s 37.5 − 25
For parallel rays, s p′ = f = 10 cm. Then, from Eq. (20 28) and with p = ′ = = 0.2
s + s 37.5 + 25
1.53

1 1 1 1
′ − = 0 + 0 + (6.5) (0.5) (0.2)2 + ⇒ sh′ = 35.68 cm
sh 37.5 cm (8)15 (1.5) (0.5 ) cm
3 0.5
bz = s p′ − sh′ = (37.5 − 35.68) cm = 1.82 cm; b y = h/s ′) bz = (2/37.5) (1.82 cm) = 0.0970 cm
20-11. Since L = (s p′ − sh′ )/(s p′ sh′ ) = bz (s p′ sh′ ), a minimum in L gives a minimum in bz. From Eq. (20 28),
h2

dL 1 2 (n + 2)
= σ + 4 (n + 1) p =0
dσ 8 f 3 n (n − 1) n−1
2 (n + 2) 2 (n2 − 1)
σ = − 4 (n + 1) p ⇒ σ = − p
n−1 n+2
20-12. The equations,

r2 + r1 1 1 1
σ= and = (n − 1) −
r2 − r1 f r1 r2
can be solved simultaneously for r1 and r2. The general result is,
2 f (n − 1) 2 f (n − 1)
r1 = and r2 =
σ +1 σ−1
For σ = 0.7, n = 1.7, and f = 30 cm: r1 = 17.65 cm and r2 = − 100 cm
For σ = 3, n = 1.7, and f = 30 cm: r1 = 7.50 cm and r2 = 15.0 cm
109
20-13. The image position is,
sf (30) (20
s′ = = cm = 60 cm
s − f 30 − 20
The Coddington shape factor is, (see problem 20 11)
2 (n2 − 1) s ′ − s 2 (1.62 − 1) 30 − 60
σ=− =− = 0.289
n+2 s +s ′ 1.6 + 2 30 + 60
Then, using the results from problem 20 12
2 f (n − 1) 2 (20) (1.6) 2 f (n − 1) 2 (20) (0.6)
r1 = = cm = 18.62 cm and r2 = = = − 33.75 cm
σ+1 1.289 σ −1 − 0.711
20-14. In either case the radius of curvature of the spherical side found from the lensmaker’s formula is ± 50 cm.
For the ideal case (see problem 20 11)
2 (n2 − 1) s ′ − s 2 (n2 − 1) s ′/s − 1 2 (1.62 − 1)
σ=− = − = − ( − 1) = 0.867
n + 2 s′ + s n + 2 s ′/s + 1 1.6 + 2
For light incident on the plane side,
r2 + r1 r2/r1 + 1
σ= = = − 1 as r1 → ∞
r2 − r1 r2/r1 − 1
For light incident on the spherical side,
r2 + r1 1 + r1/r2
σ= = = + 1 as r2 → ∞
r2 − r1 1 − r1/r2
Since σ = 1 is closer to 0.867 than is σ = − 1, the second orientation is better.
20-15. Using the result from problem 20 11,

2 (n2 − 1) s ′ − s
σ=−
n + 2 s′ + s
2 (n2 − 1) 2 (1.52 − 1)
(a) As s → ∞, σ → − ( − 1) = ( − 1) = 0.714.
n+2 1.5 + 2
(b) Using the result found in the solution to problem 20 12:
2 f (n − 1) 2 (30) (1.5 − 1) 2 f (n − 1) 2 (30) (1.5 − 1)
r1 = = cm = 17.5 cm and r2 = = cm = − 105 cm
σ+1 0.714 + 1 σ−1 0.714 − 1
2 (n2 − 1) 2 (1.52 − 1)
(c) As s ′ → ∞, σ → − (1) = (1) = − 0.714. (Reverse the lens.)
n+2 1.5 + 2
2 n2 − n − 1 s − s′ (2 )1.52 − 1.5 − 1

20-16. (a) Using Eq. (20 31) σ = = ( + 1) = 0.8
n+1 s + s′ s→∞ 1.5 + 1
(b) Using the result found in the solution to problem 20 12:
2 f (n − 1) 2 (30) (1.5 − 1) 2 f (n − 1) 2 (30) (1.5 − 1)
r1 = = cm = 16.7 cm and r2 = = cm = − 150 cm
σ+1 0.8 + 1 σ−1 0.8 − 1
(2 )1.52 − 1.5 − 1
2
s − s′

2n −n−1
(c) σ = = ( − 1) = − 0.8 (Reverse the lens)
n+1 s + s ′ s ′ →∞ 1.5 + 1
2 (n2 − 1) s ′ − s r + r1
20-17. With s ′ = s, σ = − = 0. So with σ = 2 = 0 ⇒ r1 = − r2. Then,
n + 2 s′ + s r2 − r1

1 1 1 2
= (n − 1) − = (1.5 − 1) ⇒ r1 = f = 20 cm, r2 = − r1 = − 20 cm
f r1 r2 r1
110
20-18. In this case,
2 n2 − n − 1 s − s′

σ= = 0
n+1 s + s′ s=s ′
Thus the answers are the same as in problem 20 17.
20-19. Generally,
X 1 1
=
ni f i R p
For a flat Petzval surface R p → ∞, so
1 1
+ = 0 ⇒ f2 = − (n1/n2) f1 = − (1.523/1.72) (20) cm = − 17.71 cm
n1 f 1 n2 f 2
20-20. (a) The radius of the Petzval surface is given by Eq. (20 32) as,
1 X 1 1 1
= = + ⇒ R p = 15.7 cm
Rp ni fi (1.5736) (3.543 cm) (1.6039) ( − 5.391 cm)
1 1 1
(b) =0= + ⇒ f2 = − 3.476 cm
Rp (1.5736) (3.543 cm) 1.6039) f2
20-21. The calculations proceed stepwise as follows given the values of the refractive indices from Table 20 1 and
the desired overall focal length fD of the doublet.
1 n −1 n −1
PD = , V1 = 1D , V2 = 2D
fD n1F − n1C n2F − n2C
r1 1 r2 2
2 − V1 V2 P1D P2D
P1D = PD , P2D = PD , K1 = , K2 =
1 V 2 − V1 V 2 − V 1 n 1D − 1 n 2D −1
r11 = 2/K1, r12 = − r11 , r21 = r12 , r22 = r12 /(1 − K2 r12 )
r1 2 r2 1
1 1 f f
f1D = , f2D = , fD = 1D 2D
K1 (n1D − 1) K2 (n2D − 1) f1D + f2D
The last line is repeated for fC and fF . A typical Basic program that implements this routine is given
below,
INPUT ′′N1C,N1D,N1F ? ′′;MC,MD,MF
INPUT ′′N2C,N2D,N2F ? ′′;NC,ND,NF
INPUT ′′DESIRED F ? ′′;FD
P 1/FD
V1 (MD-1)/(MF-MC):V2 (ND-1)/(NF-NC)
P1 -P*V1/(V2-V1):P2 P*V2/(V2-V1)
K1 P1/(MD-1):K2 P2/(ND-1)
R1 2/K1:R2 -R1:S1 R2:S2 R2/(1-K2*R2)
D1 1/(K1*(MD-1)):D2 1/(K2*(ND-1)):D D1*D2/(D1+D2)
C1 1/(K1*(MC-1)):C2 1/(K2*(NC-1)):C C1*C2/(C1+C2)
F1 1/(K1*(MF-1)):F2 1/(K2*(NF-1)):F F1*F2/(F1+F2)
PRINT R1;R2
PRINT S1;S2
PRINT D1;D2
PRINT D
PRINT C1;C2
PRINT C
PRINT F1;F2
PRINT F
END
The indices of refraction of the glasses in this problem are given in Table 20 1 as,
nC nD nF
#517/645 1.51461 1.51707 1.52262
#620/380 1.61564 1.62045 1.63198
problem 20 21 continued on the next page ...
111
20-21. continued...
The step by step calculations proceed as,
n1D − 1
feq = 20 cm V1 = = 64.5531
n1F − n1C
1 1 n2D − 1
PD = = D=5D V2 = = 37.9712
fin m 0.2 n2F − n2C
− V1 V2
P1D = PD − 0.121423 cm−1 P2D = PD = − 0.071423 cm−1
V2 − V1 V2 − V1
P1D P2D
K1 = = 0.234830 cm −1 K2 = = − 0.11515 cm−1
n1D − 1 n2D − 1
2
r11 = = 8.5168 cm r12 = − r11 = − 8.5168 cm = r21
K1
r12
r22 = = − 434.89 cm
1 − K2 r12
1 1
λD: f1D = = 8.23565 cm f2D = = − 14.0010 cm
K1(n1D − 1) K2(n2D − 1)
f1D f2D
fD = = 20.000 cm
f1D + f2D
1 1
λC : f1C = = 8.2750 cm f2C = = − 14.1104 cm
K1(n1C − 1) K2(n2C − 1)
f1C f2C
fC = = 20.0096 cm
f1C + f2C
1 1
λF : f1F = = 8.1428 cm f2F = = − 13.7456 cm
K1 (n1F − 1) K2 (n2F − 1)
f1F f2F
fF = = 20.0096 cm
f1F + f2F
Note: The values listed above are given rounded off in each step, the actual calculations were carried out
with the more values stored in the computer.
Problems 20 22 and 20 23 are on the next page.
112
20-22. The calculations proceed as outlined in detail in the solution to problem 20 21. Data and results are listed
here. feq = 5 cm, PD = 0.2 cm −1.
nC nD nF
#517/645 1.63461 1.63810 1.64611
#620/380 1.79608 1.80518 1.82771
(a) r11 = 3.4535 cm, r12 = − r11 = − 3.4535 cm = r21 , r22 = − 12.6576 cm
f1 (cm) f2 (cm) f (cm)

λD 2.7061 − 5.8985 5.0000
(b)
λC 2.7210 − 5.9659 5.0026
λF 2.6726 − 5.7380 5.0026
(c) P1D = 0.369534 cm−1 = 1/f1D, P2D = − 0.169534 cm−1 = 1/f2D.
1 1
Dispersive powers: ∆1 = = 0.018022, ∆2 = = 0.039283
V1 V2
(d) Check: V2 P1D + V1 P2D = (25.4562) (0.369534 cm−1) + (55.48696) ( − 0.169534 cm−1) = 0
20-23. The calculations proceed as outlined in detail in the solution to problem 20 21. Data and results are listed
here. feq = − 10 cm, PD = − 0.1 cm −1.
nC nD nF
#573/574 1.56956 1.57259 1.57953
#689/312 1.68250 1.68893 1.70462
(a) r11 = − 5.2415 cm, r12 = − r11 = − 5.2415 cm = r21 , r22 = 53.1840 cm
f1 (cm) f2 (cm) f (cm)

λD − 4.5770 8.4399 − 10.0000
(b)
λC − 4.6013 8.5194 − 10.0050
λF − 4.5222 8.2520 − 10.0050
113
Chapter 21 Fourier Optics
632.8 × 10−6 mm × 500mm

mλf
21-1. (a) m λ = d sin θ = Y m d/f or Y m = =m =0.6328 mm
d 0.5mm
For m = 1, Y1 = 0.633 mm. Note that, since d/b = 2 for the Ronchi ruling, missing orders are those for which
m = 2, 4, 6, ... Then:
m = 0, Y0 = 0 (DC component)
m = 1, Y1 = 0.633mm
m = 3, Y3 = 1.898mm
m = 5, Y5 = 3.164mm
m 1 d 0.5 mm
(b) For the fundamental frequency m = 1 so νY = = ⇒λY = = = 0.5mm.
d λY m 1

m = 1, KY = 12.57 cycles/mm
kY m 2π 2π
(c) KY = = 2 π νY = 2 π . Thus KY = m=m m = 3, KY = 37.70 cycles/mm
f d d 0.5mm m = 5, K = 62.83 cycles/mm
Y

1 2 1 1
(d) f (Y ) = + sin ω Y + sin 3 ω Y + sin 5 ω Y + ... . So the relative fundamental amplitude is 1, the
2 π 3 5
relative amplitude for the second spot is 1/3 and the relative amplitude for the third spot is 1/5. Since
irradiance is proportional to the square of the amplitude,
I2/I1 = 1/9 and I3/I1 = 1/25
21-2. (a) At any point, what is transmitted is a product: If E0 is the incident amplitude, E1 = t1E1 passes the
first transparency and E2 = t1 E1 = t2 t1 E0 passes the second transparency. Thus the combined transmission
function is t = t1 t2.
(b) The pattern in the spectrum plane would look like:
21-3. It is given that the optical density D is given by

1
D = log10 (opacity) = log10
T
1
(a) D = log10 = − log10 (T )
T
(b) As in problem 21 2, T = T1 T2 T3 . Therefore,
D = − log10 (T1 T2 T3) = D = − (log T1 + log T2 + log T3)=D1 + D2 + D3
(c) Since the opacity of the ith transparency is (opacity)i = 1.25, the optical density can be formed as
Di = log10 (1.25) = 0.09691
Dnet = 5 Di = 0.48455
Tnet = log −1 ( − D) = 10−D = 10−.48455 = 0.3277 (32.77%)
114
Et
21-4. Given T = = 5 sin(a y).
E0
(a) A bias of 5 units locates minima at T = 0.
(b)
10
-5
0 π 2π 3π 4π
No bias Bias
2
It Et
(c) = = 25 (1 + sin a y)2, For I0 = 1, It = 25 (1 + sin a y)2.
I0 E0
21-5. By definition of a Fourier transform, Z

F [h (x)] = h (x) eikx dx
Substitute for h (x) the convolution of f and g:
Z Z
F [h (x)] = eikx ′ ′ ′
f (x − x ) g (x ) d x dx
Changing the order of integration, Z

Z
′
F [h (x)] = g (x ) e ikx
f (x − x ) d x dx ′
′
Then let v ≡ x − x ′ so that dx = dv:

Z Z
′
F [h (x)] = g (x ′) eik(v+x ) f (v) dv dx ′
Z Z
F [h (x)] = f (v) e dv g (x ′) eikx dx ′ or
′
ikv
F [h (x)] = F [f (x)] × F [g (x)]
21-6. The two functions look like,
u
ρ11 (u) 6
f (x) f (x − u)
1
−3 3 −6 6
Z
Here, ρ11 (u) = f (x − u) f (x) dx. As u varies, ρ11 (u) measures the area of overlap of the square pulses:
For example,
Z 3 Z 3
2
ρ11 (0) = [f (x)] dx = dx = 6
−3
Z 3 Z−33
ρ11 (3) = f (x) f (x − 3) dx = dx = 3
−3 0
115
21-7. Consider the function y = A sin (ω t + d) = f (t). Since this function is periodic we integrate over one period
T.
Z T /2
Φ11 (τ ) = A sin (ω t + α) A sin [ω (t + τ ) + α] dt
−T /2
f (t) f (t+τ )
Z T /2 Z T /2
Φ11 (τ ) = A2 sin β sin (β + ω τ )dt = A2 sin β (sin β cos ω τ + cos β sin ω τ ) dτ
−T /2 −T /2
Z T /2 Z T /2
Φ11 (τ ) = A2 cos ω τ sin2 β dt + A2 sin ω τ sin β cos β dt
−T /2 −T /2
even o dd, int=0
Z T /2 Z T /2
Φ11 (τ ) = 2 A2 cos ω τ sin2 β d t = 2 A2 cos ω t sin2 (ω t + α) dt
0 0
T /2 T /2
2 A2 cos ω τ 2 A2 cos ω τ ω t + α sin 2 (ω t + α
Z
Φ11 (τ ) = sin (ω t + α) (ω d t) = −
ω 0 ω 2 4 0
2 A2 cos ω τ ω T sin 2 (ω T /2 + α) sin 2 α

Φ11 (τ ) = − +
ω 4 4 4
Since ω T = 2 π f T = 2 π,
2 A2 cos ω τ π sin 2 (π + α) sin 2 α

Φ11 (τ ) = − +
ω 2 4 4
 
0 1
2 A2 cos ω τ  π sin 2 π cos2 α cos 2 π sin 2 α sin 2 α 
Φ11 (τ ) = − − +
ω 2 4 4 4
cos ω τ
Φ11 (τ ) = A2 π
ω
21-8. (a) A mirror movement x ( = x2 − x1) of ∆x = λ/2 produces a change in I from one peak to the next. So the
frequency fP of the peaks will be,
∆x/∆t V 2V 2 (5 mm/s)
fP = = = = = 18.3 KHz
∆x λ/2 λ 546.1 × 10−6 mm
(b) As in measuring a wavelength difference with an interferometer, (see Eq. (8 14)), the peak to peak dis
tance is
∆x = λ2ave /2 ∆λ
v 2 v ∆λ 2 (5) (5.97 × 10−7)

fb eat = = 2 = Hz = 17.2 Hz
∆x λave (5892.94 × 10−7)2
21-9. If the total mirror movement is 5 cm, then xω = 2 × 5 = 10cm. According to Eq. (21 43), ∆λ = λ2/xw.
(6328)2 (104)2
(a) ∆λ = Å = 0.040 Å (b) ∆λ = Å = 0.100 Å
10 × 108 10 × 108
116
21-10. (a) During one pass, the total mirror movement is
nm
x = (71.5 nm/s) (time) = 71.5 [256 rdg./(1.28 rdg./s)] = 14300nm
s
xω = 2 x = 28, 600 nm = 2.86 × 103 cm
λ2 (400 × 10−7 cm)2

(b) λm in = = = 5.59 × 10−7 cm = 5.59 nm
xω 2.86 × 10−3 cm
2 xω 2 (2.86 × 10−3
(c) λmin = = = 2.24 × 10−5 cm = 224nm
N −1 255
2V 2 (71.5 nm/s)
(d) Peak frequency = = = 0.397/s. Nyquist: take at least 2 × or 0.8 readings/s.
λshortest 360nm
21-11. The given information is,
xw = 2.78mm λ: (4400 − 400)cm−1

4400cm −1 = 2.27 µm 400cm −1 = 25 µm (both infrared)
1 1
(a) We want the wave number ν wn = . For resolution we need, ∆νwn = 2 ∆λ, where ∆λ = λ2/xω. Thus,
λ λ
λ2

1 1 1
∆νwn = = = = 3.597cm−1 ≈ 3.6cm−1
λ2 xω xω 0.278cm
2 xω 2 (2.78 × 10−3 m

(b) N − 1 = = = 2449 or N = 2450
λm in 2.27 × 10−6 m
2.78mm
(c) V = = 0.0927 mm/s
30 s
117
Chapter 22 Theory of Multilayer films
22-1. Refer to the figure below for the orientation of the field vectors.
The electric field vectors are all parallel to the plane of

Film incidence and the magnetic field vectors are all into the page.
Br2
Br1 Bi1
B
The tangential components are continuous:
E
E E
B
B
θt2 Ba = B0 + Br1 = Bt1 + Bi1
θt1 B Bb = Bi2 + Br2 = Bt2
E Substrate
B
Ea = E0 cos θ0 − Er1 cosθ0 = Et1 cos θt1 − Ei1 cos θt1
E
Eb = Ei2 cos θt1 − Er2 cos θt1 = Et2 cos θt2
B0 Bt1 Bi2 Bt2
n0 n1 ns √
Using B = n ε0 µ0 E, the E equations can be expressed in
(a) (b) terms of B:
B0 cos θ0 B cos θ0 Bt1 cos θt1 Bi1 cos θt1

√ − r1√
Ea = = √ − √
n0 ε0 µ0 n0 ε0 µ0 n1 ε0 µ0 n1 ε 0 µ 0
B cos θt1 Br2 cos θt1 Bt2 cos θt2
Eb = i2√ − √ = √
n1 ε0 µ0 n1 ε0 µ0 ns ε 0 µ 0
√ √ √
n0 ε0 µ0 n1 ε 0 µ 0 ns ε0 µ0
Define γ0 ≡ , γ1 = , γs =
cos θ0 cos θt1 cos θt2
Then the pair of equations becomes,
B B B B 1 1
Ea = 0 − r1 = t1 − i1 ⇒ Ea = (B0 − Br1) = ((Bt1 − Bi1)
γ0 γ0 γ1 γ1 γ0 γ1
B B B 1 1
Eb = i2 − r2 = t2 ⇒ Eb = (Bi2 − Br2) = Bt2
γ1 γ1 γs γ1 γs
Now, Bi2 = Bt1 e−iδ and Bi1 = Br2 e−iδ . Using these in the expressions developed for Eb and Bb:
Bt1e−iδ Bi1 eiδ Bt2
Bb = Bt1 e−iδ + Bi1 eiδ = Bt2 Eb = − =
γ1 γ1 γs
B + γ1 Eb iδ B − γ1 Eb −iδ
Solving these simultaneously for for Bi1 and Bt1 give, Bt1 = b e and Bi1 = b e
2 2
Substituting these into the expressions for Ea and Ba gives,
B + γ1 Eb iδ Bb − γ1 Eb −iδ
Ba = Bt1 + Bi1 ⇒ Ba = b e + e
2 2
Bt1 Bi1 Bb + γ1 Eb iδ Bb − γ1 Eb −iδ
Ea = − ⇒ Ea = Bt1 = e − e
γ1 γ1 2 γ1 2 γ1
Simplifying using 2 cos δ = eiδ + e−iδ and 2 i sin δ = eiδ − e−iδ,
Ba = Bb cos δ + Eb (i γ1 sin δ)
i sin δ

E a = Bb + Eb cos δ
γ1
In matrix form,
    
Ea cos δ i (sin δ)/γ1 Eb
 =  
Ba i γ1 sin δ cos δ Bb
Note that this looks the same as Eq. (22 24) but in the case at hand with the electric field parallel to the
√
plane of incidence γ1 = n1 ε0 µ0 /cos θt1 while for the case treated in detail in the book leading to Eq. (22
√
24) γ1 = n1 ε0 µ0 cos θt1.
118
22-2. (a) The film thickness t should satisfy
λF λ
t= = 0
4 4 n1
√ p
with n1 = n0 ns = (1) (1.5) = 1.22. Thus
500 nm
t= = 102 nm
4 (1.22)
(b) The phase difference is

2π 2π
δ = k0 ∆ = (n1 t cos θt1) = (1.22) (102 nm) (1) = 1.428 rad
λ0 550 nm
So, cos δ = 0.1423 and sin δ=0.9898. Then,
n21 (n0 − ns)2 cos2δ + (n0 ns − n21)2 sin2δ
R= with n0 ns − n21 = 0
n21 (n0 + ns)2 cos2δ + (n0 ns + n21)2 sin2δ
1.5 (1 − 1.5)2 (0.142315)2 + 0

R=
1.5 (1 + 1.5)2 (0.142315)2 + (1.5 + 1.5)2 (0.989821)2
R = 8.432 × 10−4 ⇒ 0.0843%
22-3. From Eq. (22 42) or the relation used in problem 22 2, with δ = 2 π ∆/λ = 2 π (λ/2)/λ = π, so that cos2 δ = 1
and sin2 δ = 0,
n2 (n − ns)2 (1) + 0 (n0 − ns)2
R = 12 0 =
n1 (n0 + ns)2 (1) + 0 (n0 + ns)2
22-4. Using Eq. (22 4),

n21 (n0 − ns)2 cos2δ + (n0 ns − n21)2 sin2δ 1.462 (1 − 1.52)2 cos2 δ + (1.52 − 1.462)2 sin2 δ
R= =
n21 (n0 + ns)2 cos2δ + (n0 ns + n21)2 sin2δ 1.462 (1 + 1.52)2 cos2 δ + (1.52 + 1.462)2 sin2 δ
2π 2π 2π
(a) For λ = 800 nm, δ = ∆= (n1 t) = (1.46 )(137 nm) = π/2. So,
λ0 λ0 800 nm
∆/λ = 200/800 = 1/4, cos δ = 0, sin δ = 1
R = 0.0281 ⇒ R = 2.81%
2π 2π ∆ 200 1
(b) For λ = 600 nm, δ = 200 = . So, = = .
600 3 λ 600 3
cos δ = − 0.5, sin δ = 0.866
R = 0.0317 ⇒ R = 3.17%
2π ∆ 200 1
(c) For λ = 400 nm, δ = 200 = π. So, = = .
400 λ 400 2
cos δ = − 1, sin δ = 0
R = 0.0426 ⇒ R = 4.26%
(Same as glass of n = 1.52)
22-5. The phase difference is δ = (2 π/λ0) (n1 t) = (2 π/5600) (2.35)(596) = π/2 (∆ = λ/4). This is a quarter wave
plate at normal incidence so Eq. (22 43) may be used,
2 2
n0 ns − n21 (1)(1.52) − (2.35)2

R= = = 0.323 ⇒ R = 32.3%
n0 ns + n21 (1)(1.52) + (2.35)2
119
22-6. The properties of the film for zero reflectance are,
√ p
n1 = n0 ns = (1) (4) = 2
λ 2 µm
t= 0 = = 0.25 µm
4 n1 4 (2)
According to Table 22 1, ZrO2 has an index of refraction of 2.0 in the near infrared and so could be used as
the film.
22-7. The double layer arrangement is as shown.
(a) The layers have quarter wave thickness for λ0 = 550 nm:
n0 = 1 λ0 λ0 550 nm λ0 550 nm
t1 = = = 105.8 nm, t2 = = = 85.94 nm
n1 = 1.30 4 n1 4 (1.3) 4 n2 4 (1.6)
n2 = 1.60
From Eq. (22 45):
ns = 1.52
2 2
n0 n22 − ns n21 (1) (1.6)2 − 1.52 (1.3)2

R= = = 3 × 10−6 ⇒ R = 0.0003%
n0 n22 + ns n21 (1) (1.6)2 + 1.52 (1.3)2
(b) With the layers reversed:
2 2
n0 n22 − ns n21 (1) (1.3)2 − 1.52 (1.6)2

R= = = 0.1555 ⇒ R = 15.55%
n0 n22 + ns n21 (1) (1.3)2 + 1.52 (1.6)2
22-8. The double layer arrangement is as shown.
(a) The layers have quarter wave thickness for λ0 = 550 nm:
n0 = 1 λ0 = 2µm
n1 = 1.35 λ0 2000
n2 = 2.2 t1 = = nm = 370 nm
4 n1 4 (1.35)
ns = 3.3 λ 2000 nm
t2 = 0 = = 227 nm
4 n2 4 (2.2)
r
3.3
r
n2 ns
(b) The ideal index ratio is = = = 1.8166
n1 n0 1
1.8166 − 1.6296
The actual ratio is n2/n1 = 2.2/1.35 = 1.6296. The % difference is 100% = 10.3%
1.8166
(c) The normal reflectance produced is
2 2
n0 n22 − ns n21 (1) (2.2)2 − 3.3 (1.35)2

R= = = 0.0117 ⇒ R = 1.17%
n0 n22 + ns n21 (1) (2.2)2 + 3.3 (1.35)2
For the λ/4 layer: cos δ1 = cos (π/2) = 0, sin δ1 = sin (π/2) = 1.
For the λ/2 layer: cos δ2 = cos (π) = − 1, sin δ2 = sin (π) = 0.
The transfer matrix is

       
cos δ1 i sin δ1/γ1 cos δ2 i sin δ2/γ2 0 i /γ1 −1 0 0 − i /γ1
M =  =  = 
i γ1 sin δ1 cos δ1 i γ2 sin δ2 cos δ2 i γ1 0 0 −1 − i γ1 0
λ/4 λ/2
Thus the matrix elements of the overall transfer matrix are the negatives of those for the λ/4 layer acting
alone. Thus the reflection coefficient given in Eq. (22 36) for the double layer and the single λ/4 layer differ
by only an overall minus sign. Since R = |r|2, the reflectances of the two layer structures is the same. Note
that this is only true at the wavelength λ0 for which the layers are λ/4 and λ/2.
120
22-10. For a double layer of arbitrary thickness, (to be specified), but normal incidence, the transfer matrix is
  
cos δ1 i sin δ1/γ1 cos δ2 i sin δ2/γ2
M = M1 M2 =  
i γ1 sin δ1 cos δ1 i γ2 sin δ2 cos δ2
   
cos δ1 cos δ2 − (γ2/γ1) sin δ1 sin δ2 i(cos δ1 sin δ2/γ2 + sin δ1 cos δ2/γ1) m11 m12
M = ≡ 
i (γ2 cos δ1 sin δ2 + γ1 sin δ1 cos δ2) cos δ1 cos δ2 − (γ1/γ2) sin δ1 sin δ2 m21 m22
The general form of the reflection coefficient is, (since in our case m12 and m21 are pure imaginary)
γ m + γ0 γs m12 − m21 − γs m22
r = 0 11
γ0 m11 + γ0 γsm12 + m21 + γs m22
Now, let us put this result in a form convenient for programming. Since in our case m12 and m21 are pure
imaginary, write m12 = i µ12 , m21 = i µ21 where µ12 and µ21 are real. Then the reflection coefficient can be
recast as,
(γ m − γs m22 ) + i (γ0 γs µ12 − µ21 )
r = 0 11
(γ0 m11 + γs m22 ) + i (γ0 γs µ12 + µ21 )
Making the substitutions,
√ √ √ √
γ0 = n0 ε0 µ0 , γ1 = n1 ε0 µ0 , γ2 = n2 ε0 µ0 , and γs = ns ε0 µ0 in this expression (including in the matrix
elements) leads to
(n0 M − ns P ) + i (n0 ns N − O) A + i B
r= ≡
n0 M + ns P ) + i (n0 ns N + O) C +iF
Here
M = cos δ1 cos δ2 − (n2/n1) sin δ1 sin δ2 P = cos δ1 cos δ2 − (n2/n1) sin δ1 sin δ2
N = cos δ1 sin δ2/n2 + sin δ1 cos δ2/n1 O = n2 cos δ1 sin δ2 + n1 sin δ1 cos δ2
The reflectance is,
2 A2 + B 2
R = |r| =
C2 + F 2
2π
To calculate M , N , O, and P , the phase differences must be formed. For a given wavelength λ, δ = n t,
λ0
where t is the film thickness. Write, t = Q (λ0/n) where, for example, Q = 1/4 for a λ/4 layer. Q will be
variable to account for different film thicknesses. We will do this for each layer. In addition to the variables
M , N , O, P and A, B, C , F defined above the programming variables are:
n1 → G n2 → H ns → I
λ0 → L t1 → T = Q λ0/n1 t2 → U = S λ0/n2
λ0 → W ( in loop) δ1 → D δ2 → E
A program to calculate the reflectance for different parameters is given below:
5 PRINT ′′2-LAYER FILM ′′

10 INPUT ′′N1 ′′;G, ′′N2 ′′;H, ′′NS ′′;I, ′′WL0(MICRON) ′′;L
15 INPUT ′′THICK1 ′′;Q, ′′THICK2 ′′;S

20 T Q*L/G:U S*L/H
25 F0R W 0.3 TO 0.8 STEP 0.05
30 D 2*PI*G*T/W:E 2*PI*H*U/W
35 M COS(D)*COS(E)-H*SIN(D)*SIN(E)/G
40 N COS(D)*SIN(E)/H+SIN(D)*COS(E)/G
45 O G*SIN(D)*COS(E)+H*COS(D)*SIN(E)
50 P COS(D)*COS(E)-G*SIN(D)*SIN(E)/H
55 A M-I*P:B I*N-O:C M+I*P:F I*N+O
60 R 100*((A*A+B*B)/(C*C+F*F))
65 PRINT ′′WL ′′;W, ′′REF ′′;R
70 NEXT W
75 END
Sample output is given on the next page.
121
22-10. Continued...
To generate Figure 22 4 in the text, The program output is:
use the following input data:
λ µm Ra Rb Rc
Case a Case b Case c 0.3 7.84% 4.91% 7.96%
n1 = 1.65 n1 = 1.38 n1 = 1.38 0.35 16.19 4.48 10.63
n2 = 2.1 n2 = 1.6 n2 = 1.85 0.40 14.02 1.80 3.11
Q = 0.25 Q = 0.25 Q = 0.25 0.45 7.24 0.90 0.013
S = 0.25 S = 0.25 S = 0.4 0.5 1.84 1.11 0.74
In all cases ns = 1.52 and λ0 = 0.55 µm. 0.55 0.10 1.26 1.26
0.60 1.32 1.15 0.88
0.65 3.95 0.97 0.26
0.70 6.84 0.90 0.0032
0.75 9.45 0.98 0.32
0.80 11.6 1.22 1.16
22-11. Quarter wave matrices (as in the solution to problem 22 9) for normal incidence give a composite transfer
matrix of
       
0 i /γ1 0 i /γ2 0 i /γ3 0 − iγ2 /(γ1 γ3) m11 m12
M =    =  ≡ 
i γ1 0 i γ2 0 i γ3 0 − iγ1 γ3/γ2 0 m21 m22
Using Eq. (22 36),

γ0 m11 + γ0 γs m12 − m21 − γs m22
r=
γ0 m11 + γ0 γsm12 + m21 + γs m22
1 3 γ γ γ γ γ
γ0 γs [ − i γ2/(γ1 γ3)] + i γ1 γ3/γ2 γ2
− 0γ sγ 2
1 3
r= =
γ0 γs [ − i γ2/(γ1 γ3)] − i γ1 γ3/γ2 − γ1γ γ3 − γ0γ γsγγ2
2 1 3
Now r = 0 when
γ1 γ3 γ0 γs γ2
= ⇒ γ0 γs γ22 = γ12 γ32
γ2 γ1 γ3
√
Since for each case γ = m ε0 µ0 . Substituting this for each γ and simplifying gives
n0 ns n22 = n21 n23
n1 n3 √
= n0 ns
n2
22-12. We need , for R = 0,

n1 n3 √ √
= n0 ns = 1 × 4 = 2
n2
Using the data in Table 22 1, we see that a good approximations can be formed by using MgF2 for layer 3
so that n3 = 1.35, SiO for layer 2 so that n2 = 1.5, and ZnS for layer 1 so that n1 = 2.2. For these choices,
n1 n3 (2.2)(1.35)
= = 1.98 ≈ 2
n2 1.5
22-13. Using Eq. (22 53),

 2  2N 2
1.38 2N

n0 nL 1
ns
− 1   1.52
nH 2.6
− 1
R = 2N  =
 
1.38 2N

n0 nL 1
n n
+ 1 1.52 2.6
+ 1
s H
(a) N = 2, R = 81.1% (b) N = 4, R = 98.4% (c) N = 8, R = 99.99%
122
22-14. Substituting in Eq. (22 53) as in problem 22 13, one finds for N = 4, nH = 4, nL = 1.35, and ns = 1.50.
22-15. In Eq. (22 53), let x = n0/ns and z = nL/nH . Then
√ 1 − x z 2N
± R=
1 + x z 2N
since x < 1 and z < 1 we select the positive square root by writing,
√ 1 − x z 2N
+ R=
1 + x z 2N
Then,
√ √
R + x z 2N R = 1 − x z 2N
√ √
x z 2N 1 + R = 1 − R
√
1− R
z 2N = p
x (1 + R)
In the case at hand, this becomes,

√
1 − 0.9
4
z = √ = 0.03950
(1/1.5) (1 + 0.9 )
z = 0.4458
nH 1
= = 2.243
nL z
22-16. As in the solution to problem 22 15, with x = n0/ns and z = nL/nH ,

2
x z 2N − 1

R=
x z 2N + 1
2
−1
As N → ∞, z 2N
→ 0 since z < 1. In this case R → =1
1
2
−1
As z → 0, R → = 1.
1
123
Chapter 23 Fresnel equations
23-1. From Eq. (23 28), √

− n2 cos θ + n2 − sin2 θ
rT M = √
n2cos θ + n2 − sin2θ
The numerator vanishes when,
√
n2 cos θ = n2 − sin2 θ
Squaring gives,
n4 cos2 θ = n2 − sin2 θ ⇒ n4 cos2θ − n2 + sin2θ = 0
This is a quadratic in n2,
√ √ p
2 + 1 ± 1 − 4 sin2θ cos2θ 1 ± 1 − 4 sin2θ + 4 sin4 θ 1 ± (2 sin2θ − 1)2
n = = =
2 cos2θ 2 cos2 θ 2 cos2 θ
1 ± (2 sin2 θ − 1)
n2 =
2 cos2 θ
sin2 θ
+ sign : n2 = = tan2 θ ⇒ n = tan θ Brewster ’ s law
cos2 θ
2 (1 − sin2θ)
− sign: n2 = =1 Trivial Solution
2 cos2 θ
23-2. The critical angle is,

n1 1 1 1
sin θc = = ⇒n= = = 1.8094
n2 n sin θc sin (33◦33 ′)
External Reflection:
n2
tan θ p = = n ⇒ θp = tan−1(1.8094) = 61◦4 ′
n1
Internal Reflection:
n 1 1
tan θ p′ = 1 = ⇒ θ p′ = tan−1 = 28◦56 ′
n2 n 1.8094
23-3. The critical angle is,

θc = sin−1(1/n) = sin−1(1/1.84) = 32.9◦
The polarizing angles are,
External: θ p = tan−1(n) = tan−1 1.84) = 61.5◦
′
Internal: θ p = tan−1(1/n) = tan−1(1/1.84) = 28.5◦
23-4. Given,
1
sin θc ≡ sin θ = and tan θ p ≡ tan θ = n
n
Then,
sin θ 1/n
tan θ = =p =n
cos θ 1 − (1/n)2
1 p
= 1 − (1/n)2
n2
1
= 1 − (1/n)2
n4
1 = n4 − n2 ⇒ n4 − n2 − 1 = 0
√ √ √
1± 1+4 1+ 5 1− 5
2
n = = , Take positive value
√2 2 2
1+ 5
n2 = ⇒ n = 1.272 = n2/n1 = n2
2
124
23-5. In each of the 4 equations the desired form results from introducing θt through the relation n = sin θ/sin θt.
For example:
√
q
cos θ − sin θ cos2θt)/sin2 θt
p
cos θ − n − sin θ cos θ − sin θ 1/sin θt − 1
2 2 2
rT E = √ = =
cos θ + n2 − sin 2 θ cos θ + sin θ 1/sin2 θt − 1 cos θ + sin θ
p q
cos2θt)/sin2 θt
cos θ sin θt − sin θ cos θt sin(θ − θt)
rTE = =−
cos θ sin θt + sin θ cos θt sin(θ + θt)
The other 4 relations are similarly shown to be true. Proceeding by first noting from above that,
√
n2 − sin2θ = sin θ cos θt/sin θt
Then,
√ sin 2 θ cos θ sin θ cos θt
− n2 cos θ + n2 − sin2 θ − sin 2 θt + sin θt
rT M = √ = sin 2 θ cos θ sin θ cos θ
n2 cos θ + n2 − sin2 θ + sin θ t
sin 2 θ t t
1 1
− sin θ cos θ + cos θt sin θt − sin θ cos θ(sin2 θt + cos2 θt) + cos θt sin θt sin2 θ + cos2 θ

rT M = =
sin θ cos θ + cos θt sin θt sin θ cos θ(sin2 θt + cos2 θt) + cos θt sin θt (sin2 θ + cos2 θ )
− sin θ cos θ sin 2 θt − sin θ cos θ cos2 θt + cos θt sin θt sin2 θ + cos θt sin θt cos2 θ
rT M =
+ sin θ cos θ sin 2 θt + sin θ cos θ cos2 θt + cos θt sin θt sin2 θ + cos θt sin θt cos2 θ
(sin θ cos θt − cos θ sin θt) (sin θ sin θt − cos θ cos θt) sin(θ − θt)cos(θ + θt)
rTM = =−
(sin θ cos θt + cos θ sin θt) (sin θ sin θt + cos θ cos θt) sin(θ + θt)cos(θ − θt)
tan(θ − θt)
rT M = −
tan(θ + θt)
and,
2 cos θ 2 cos θ 2 cos θ sin θt 2 cos θ sin θt
tT E = √ = = =
cos θ + n2 − sin2θ cos θ + sin θ cos θt/sin θt cos θ sin θt + sin θ cos θt sin (θ + θt)
Finally,
2 n cos θ 2 cos θ sin θ/sin θt 2 cos θ sin θt
tT M = √ = 2 =
n2 cos2 θ + n2 − sin2θ sin θ2cos θ + sin θ cos θt cos θ sinθ + sin θt cos θt
sin θt sin θt
The denominator in this expression is the same as the one that appeared in the work leading to the final
form of rT M . Using the result derived there,
2 cos θ sin θt
tT M =
sin(θ + θt)cos(θ − θt)
23-6. Write,
√ √ −1
− n2 cos θ + i sin2 θ − n2 − a + i b a−ib a2 + b2 e−i tan (b/a)
rT M = √ ≡ =− =− √
n2 cos θ + i sin2 θ − n2 a+ib a+ib −1
a2 + b2 ei tan (b/a)
−1 −1
rT M = (1) eiΦT M = − 1 e−2itan (b/a)
= e−2itan (b/a)+iπ
ΦT M = − 2 tan−1(b/a) + π
π b √sin2 θ − n2
tan − ΦT M = =
2 a n2 cos θ
√
π sin2 θ − n2
tan ΦTM − =−
2 n2 cos θ
125
23-7. The plots below are for diamond (n = 2.42). They are produced with the free computer algebra system
Gnuplot under the TeXmacs front end.
gnuplot] n1 2.42~set title "External Reflection"~k pi/180~set xrange [0:90]~set xzeroaxis lt 1~
set size 0.7,0.7~set xlabel "{/Symbol q}" font "Times,20"
plot (cos(k*x)-sqrt(n1**2-sin(k*x)*sin(k*x)))/(cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x))) title "{/Italic
r_{TE}",(-n1*n1*cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x)))/(n1*n1*cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x)))
title "{/Italic r_{TM}",2*cos(k*x)/(cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x))) title "{/Italic
t_{TE}",2*n1*cos(k*x)/(n1*n1*cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x))) lw 3 title "{/Italic t_{TM}"
External Reflection
1
rTE
0.8 rTM
tTE
0.6 tTM
0.4
0.2
-0.2
-0.4
-0.6
-0.8
-1
0 10 20 30 40 50 60 70 80 90
θ
gnuplot] n1 1/2.42~k pi/180~set xrange [0:30]~set xzeroaxis lt 1~set samples 10000
set size 0.7,0.7~set xlabel "{/Symbol q}" font "Times,20"~set title "Internal Reflection"
plot (cos(k*x)-sqrt(n1**2-sin(k*x)*sin(k*x)))/(cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x))) title "{/Italic
r_{TE}",(-n1*n1*cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x)))/(n1*n1*cos(k*x)+sqrt(n1**2-sin(k*x)*sin(k*x)))
title "{/Italic r_{TM}"
Internal Reflection
1
rTE
0.8 rTM
0.6
0.4
02
0
-0 2
-0.4
-0.6
-0.8
-1
0 5 10 15 20 25 30
θ
23-8. The transmittance is T = 1 − R. With Gnuplot scripts similar to those of the last solution,
1
09
0.8
0.7
0.6
0.5
T
0.4
03
0.2
0.1
0
0 10 20 30 40 50 60 70 80 90
θ
Ext, TE Int, TE
Ext, TM Int, TM
126
23-9. For n = 1/1.3:
θc = sin−1(1/1.3) = 0.878 rad = 50.3◦, θ p′ = tan−1(1/1.3) = 0.6557 = 37.6◦.
For n = 1/2.42:
θc = sin−1(1/2.42) = 0.426 = 24.41◦, θp′ = tan−1(1/2.42) = 0.392 = 22.46◦.
Then plotting directly the expression in Eqs. (23 38) and (23 39) for these two cases. The Gnuplot script is
shown.
GNUplot] k pi/180~n1 1/2.42~set xrange [0:90]~jp 22.46~jc 24.41~set samples 100000~
set size 0.5,0.5~set xlabel "{/Symbol q}" font "Times,18"
set title "{/Italic n} 1/2.42"
set ytics ("-{/Symbol p}" -pi,"-{/Symbol p}/2" -pi/2,"0" 0,"{/Symbol p/2}" pi/2,"{/Symbol p}" pi )~set
xzeroaxis lt 3
f1(x) 0< x && x<jp ? 0:1/0
f2(x) jp< x && x<jc ? pi:1/0
f3(x) jc< x && x<90 ? -2*atan(sqrt(sin(k*x)*sin(k*x)-n1**2)/n1/n1/cos(k*x))+pi:1/0
g1(x) 0< x && x<jc ? 0:1/0
g2(x) jc< x && x<90 ? -2*atan(sqrt(sin(k*x)*sin(k*x)-n1**2)/cos(k*x)):1/0
plot f1(x) lw 2 notitle, f2(x) lt 1 lw 6 notitle, f3(x) lt 1 lw 2 title "{/Symbol F}_{TM}", g1(x) lt 2
lw 2 notitle, g2(x) lt 2 lw 2 title "{/Symbol F}_{TE}"
n=1/1.3 n=1/2.42
π π
ΦTM ΦTM
ΦTE ΦTE
π/2 π/2
0 0
-π/2 -π/2
-π -π
0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90
θ θ
23-10. The reflectance is

2
1−n
R=
1+n
(a) Air film: n = 1.38/1, R = 0.0255 ⇒ 2.55%
(b) Film glass: n = 1.52/1.38, R = 0.00233 ⇒ 0.233%
(c) Air Glass: n = 1.52/1, R = 0.0426 ⇒ 4.26
(d) Using Eq. (22 43)

2 2
n0 ns − n21 1.52 − 1.382

R= = = 0.0126 = 1.26%
n0 ns + n21 1.52 + 1.382
23-11. Using
√ !2 √ !2
cos θ − n2 − sin2 θ − n2 cos θ + n2 − sin2 θ
RTE = √ RT M = √
cos θ − n2 − sin2 θ n2 cos θ + n2 − sin2 θ
θ RT E RTM
0◦ 2.01% 2.01%
10◦ 2.10% 1.91%
45◦ 5.23% 0.274%
90◦ 100% 100%
127
23-12. For diamond n = 2.42.
(a) External:
TM: θ p = tan−1(n) = tan−1(2.42) = 67.55 , no critical angle.
◦
TE: No Brewster angle, no critical angle.
(b) Internal:
TM: θc = sin−1(1/2.42) = 0.426 = 24.41◦, θ p′ = tan−1(1/2.42) = 0.392 = 22.46◦.
TE: θc = sin−1(1/2.42) = 0.426 = 24.41◦, no Brewster angle.
23-13. The reflectance and transmission are given by

√ !2 p !2
cos θ − n2 − sin2 θ cos (50◦) − 1.62 − sin2(50◦)
RTE = √ = = 0.1385 = 13.85%,
cos θ − n2 − sin2 θ cos (50◦) + 1.62 − sin2(50◦)
p
TTE = 1 − RT E = 86.15%
√ !2 p !2
− n2 cos θ + n2 − sin2 θ − 1.62 cos (50◦) + 1.62 − sin2(50◦)
RTM = √ = = 0.00623 = 0.623%
n2 cos θ + n2 − sin2 θ 1.62 cos (50◦) + 1.62 − sin2(50◦)
p
TT M = 1 − RT M = 99.38%
23-14. (a) For the TE case begin with

E + Er = Et (23 20)
n1 E cos θ − n1 Er cos θ = n2Et cos θt (23 21)
Solve Eq. (23 20) for Er and use the result in Eq. (23 21),
n1 E cos θ − n1 (Et − E) cos θ = n2 Et cos θt
Rearrangement gives,
Et 2 n1 cos θ 2 cos θ 2 cos θ
tT E = = = = √
E n1 cos θ + n2 cos θt cos θ + n cos θt cos θ + n2 − sin2θ
Similarly for the TM mode

− n1 E + n1 Er = − n2 Et (23 22)
E cos θ + Er cos θ = Et cos θt (23 23)
Substitution of Er from Eq. (23 22) into Eq. (23 33) and subsequent rearrangement gives,
E cos θ + E − n Et) cos θ = Et cos θt
Et 2 cos θ 2 n cos θ 2 n cos θ
tTM = = = = √
E n cos θ + cos θt n2 cos θ + n cos θt n2 cos θ + n2 − sin2θ
(b) From Eq. (23 20), after dividing through by E, one finds
1 + rTE = tTE
√ √ √
cos θ − n2 − sin2 θ cos θ + n2 − sin2 θ + cos θ − n2 − sin2 θ 2 cos θ
tT E = 1 + √ = √ = √
cos θ + n − sin θ
2 2 cos θ + n − sin θ
2 2 cos θ + n2 − sin2θ
Similarly after dividing Eq. (23 22) by E and by n1,
− 1 + rTM = − n tT M
 √ √ 
1 1 n cos 2
θ + n 2
− sin 2
θ − − n 2
cos θ + n 2
− sin 2
θ 2 n cos θ
tTM = (1 − rT M ) =  √ = √
n n n cos θ + n − sin θ
2 2 2 n cos θ + n2 − sin2 θ
2
128
23-15. (a) The critical angle is θc = sin−1(n2/n1) = sin−1(1/1.458) = 43.3◦
The polarizing angle for external reflection is is θ p = tan−1(n2/n1) = tan−1(1.458/1) = 55.6◦
The polarizing angle for internal reflection is θ p′ = tan−1 1/1.458) = 34.4◦

2 2
1 − 1.14582

2 1−n
(b) At normal incidence: RTE = rTE = = = 0.0347, TTE = 1 − RTE = 0.9653
1+n 1 + 1.14582
√ !2 p !2
cos θ − n2 − sin2 θ cos 45◦ − (1.458)2 − sin2 45◦
At θ = 45◦: RTE = √ = = 0.0821
cos θ + n2 − sin2 θ
p
cos 45◦ + (1.458)2 − sin2 45◦
TTE = 1 − RTE = 0.9179.
(c) At normal incidence: RTM = RTE = 0.0347, TTE = 0.9653
√ !2 √ !2
− n2 cos θ + n2 − sin2 θ − (1.458)2 cos 45◦ + 1.4582 − sin2 45◦
At θ = 45◦: RTM = √ = √
n2 cos θ + n2 − sin2 θ (1.458)2 cos 45◦ + 1.4582 − sin2 45◦
RTM = 0.0067, TTM = 0.9933
(d) Use Eqs. (23 28) and (23 29):
For incident angles less than θ p′ , φTM = φTE = 0. So for θ = 0◦ and 20◦, φTM − φTE = 0.
For θ = 40◦ which lies between θ p′ and θc, φTM − φTE = π − 0 = π.
For θ > θc: with n = 1/1.458,

√ ! √ !
sin2 θ − n2 sin2 θ − n2
φTM − φTE = − 2 tan−1 + π + 2 tan−1
n2 cos θ cos θ
θ = 50◦: φTM − φTE = 2.43 rad = 139◦
θ = 70◦ φTM − φTE = 2.65 rad = 152◦
θ = 90◦ φTM − φTE = π = 180◦
23-16. (a) By trial and error, θ = θrhomb = 59.857◦, so that with n = 1/1.65
√ ! √ !
sin2 θ − n2 sin2 θ − n2
2 (φTE − φTM ) = − 2 tan −1
+ 2 tan −1
− π = − 1.5 π ⇒ π/2
cos θ n2 cos θ
(b) If θ = (1 + 0.05) 59.857◦ = 62.850◦ :

√ ! √ !
sin2 θ − n2 sin 2
θ − n 2
2 (φTE − φTM ) = − 2 tan−1 + 2 tan−1 − π = − 1.543 π ⇒ 0.457 π
cos θ n2 cos θ
If θ = (1 − 0.05) 59.857◦ = 56.864◦ :

√ ! √ !
sin2 θ − n2 sin 2
θ − n 2
2 (φTE − φTM ) = − 2 tan−1 + 2 tan−1 − π = − 1.461 π ⇒ 0. 539 π
cos θ n2 cos θ
129
23-17. Using Eq. (23 65) for the TE case:
2
q
cos θ − 2 2
nR − nI − sin θ + i (2 nR nI
2

RTE =

q
cos θ − n2R − n2I − sin2 θ + i (2 nR nI

and for the TM case,
− n2 − n2 + i (2 n n ) cos θ + n2 − n2 − sin2 θ + i (2 n n 2

R I R I R I R I
RTM =
n2R − n2I + i (2 nR nI ) cos θ + n2R − n2I − sin2 θ + i (2 nR nI

The given parameters are nR = 2.485 and nI = 1.381. The free computer algebra system Maxima with the
TeXmacs front end, can be used to determine the reflectances. Below I show the syntax used and give a
table with the output.
(%i1) (nR:2.484,nI:1.381,k:%pi/180)$
(%i2) for j from 0 step 10 thru 90 do (RTE[j]:float(abs((cos(k*j)-sqrt(nR^2-nI^2-
sin(k*j)*sin(k*j)+%i*2*nR*nI))/(cos(k*j)+sqrt(nR^2-nI^2-
sin(k*j)*sin(k*j)+%i*2*nR*nI)))^2),RTM[j]:float(abs(((-nR^2-nI^2+2*%i*nR*nI)*cos(k*j)+sqrt(nR^2-nI^2-
sin(k*j)*sin(k*j)+%i*2*nR*nI))/((nR^2-nI^2+2*%i*nR*nI)*cos(k*j)+sqrt(nR^2-nI^2-
sin(k*j)*sin(k*j)+%i*2*nR*nI)))^2))
(%o3) done
(%i4) for j from 0 step 10 thru 90 do display(RTE[j],RTM[j])
θ 0◦ 30◦ 50◦ 70◦ 90◦

RT E 29.3 34.5 45.4 65.7 100
RTM 29.3 24.2 14.9 5.4 100
23-18. Using the Maxima program from the last problem with the input values nR = 1.5 and nI = 5.3 I find,
θ RT E RTM
0◦ 82.5% 82.5%
30◦ 84.7% 80.1%
60◦ 90.9% 69.5%
23-19. Given nI = 5.3 at λ = 589.3 nm. (a) α = 4 π nI /λ = 4 π (5.3)/(589.3 nm) = 0.113 nm−1
(b) I = I0 e−αs. For I = 0.01 I0, e−αs = 0.01 ⇒ s = ( − 1/α) ln(0.01) = 40.75 nm = 0.069 λ
23-20. Consider Eq. (23 47)

cos θt

1 = r2 + n t2 n = n2/n1
cos θ
cos θt 2 2

(a) External Reflection: n > 1; θt < θi; cos θt > cos θ. Thus, 1 − r2 = nt > t ⇒ t2 < 1 − r 2
cos θ
Since r 2 < 1, t2 < 1.
cos θt 2 2

Internal Reflection: n < 1; θt > θi; cos θt < cos θ. Thus, 1 − r = n
2
t < t ⇒ t2 > 1 − r 2
cos θ
No upper limit is imposed.
(c) For the angle of incidence equal to the critical angle, sin θ = sin θc = n. Then,
2 cos θ 2 cos θ
tT E = √ = =2
cos θ +n2 − sin2 θ cos θ
2 n cos θ 2
tT M = √ =
n cos θ + n − sin θ
2 2 2 n
Continued on the next page...
130
23-20. Continued...
(c)
Internal Reflection
3
tTE
2.8 tTM
2.6
2.4
2.2
2
1.8
1.6
1.4
1.2
1
0 10 20 30 40 50
θ
23-21. (a) The penetration depth is
1 λ 1 0.546 µm 1
|z |1/e = = = = 0.164 µm
α 2 π sin2 θ/n2 − 1 2π sin (45 )/(1/1.6)2 − 1
p p
2 ◦
1
(b) Since irradiance is proportional to the square of the field amplitude and with α = = 6.089 µm−1,
|z |1/e
I
= e−2α |z| = e−2(6.089 µm 1µm )
−1
= 5.1 × 10−6
I0
131
Chapter 24 Nonlinear optics and the modulation of light
24-1. For a single field,

1
E = E0 (eiωt + e−iωt) = E0 cos ω t
2
The third order polarization is
3
E03 i3ωt ε 0 χ3 E 0
P3 = ε0 χ3 E 3 = ε0 χ3 e + e−i3ωt + 3 eiωt + 3 e−iωt = (cos 3 ω t + 3 cos ω t)
8 4
Frequencies ω and 3 ω appear.
24-2. The total electric field is

1 1
E= E01 (eiω1 t + e−iω1 t) + E02 (eiω2 t + e−iω2 t) ≡ A + B
2 2
P3 = ε0 χ3 E 3 = ε0 χ3 (A3 + B 3 + 3 A2 B + 3 A B 2)
Here,
3 E3
E01
A3 = ei3ω1 t + e−i3ω1 t + 3 eiω1 t + 3 e−iω1 t = 01 (cos 3 ω1 t + 3 cos ω1t)
8 4
3
E
B 3 = 02 [cos 3 ω2 t + 3 cos ω2 t]
8
2
3 E 01 E 02
3 A2 B = e 1 + e−i2ω1 t eiω2 t + e−iω2 t
i2ω t
8
2
3E01 E02 h i(2ω1 +w2)t i
3 A2 B = e + e−i(2ω1 +ω2)t + ei(2ω1 −w2)t + e−i(2ω1 −ω2) t + 2 (eiω2 t + e−iω2 t)
8
2
3 E01 E02
3 A2 B = [cos 2 ω1 + ω2) t + cos (2 ω1 − w2) t + 2 cos ω2 t]
4
Similarly, interchanging subscripts 1 and 2,
2
3 E02 E01
3 A B2 = [cos (2 ω2 + ω1) t + cos (2 ω2 − ω1) t + 2 cos ω1 t]
4
Frequencies appearing: 3 ω1, 3 ω2, 2 ω1 + ω2, 2 ω1 − ω2, 2 ω2 + ω1, 2 ω2 − ω1, ω1, ω2 .
24-3. The total electric field is,

E = E01 cos ω1 t + E02 cosω2 t + E03 cos ω3 t
E01 iω1 t E E
+ e−iω1 t + 02 eiω2 t + e−iω2 t + 03 eiω3 t + e−iω3 t

E= e
2 2 2
The second order polarization is P2 = ε0 χ2 E 2. Algebra similar to that of the previous two problems leads
to
ε χ n 2
P2 = 0 2 (E01 2
+ E02 + E03 2 2
) + E01 2
cos 2 ω1 t + E02 2
cos 2 ω2 t + E03 cos 2 ω3 t +
2
+ 2 E01 E02 [cos (ω1 + ω2) t + cos (ω1 − ω2) t + cos (ω2 + ω3) t + cos (ω2 − ω3) t + cos (ω1 + ω3) t + cos (ω1 − ω3) t}
Frequencies include: 0, 2 ω1, 2 ω2, 2 ω3, ω1 ± ω2, ω1 ± ω3, ω2 ± ω3
24-4. For crystals with inversion symmetry, 1/n2 cannot change by reversing E. From Eq. (24 7):
1 1
For + E: = + r ( + E) + R ( + E)2
n2 n20
1 1
For − E: 2 = 2 + r ( − E) + R( − E)2
n n0
For these to be equal for all E, r E = − r E ⇒ r = 0.
132
λ0 694 nm λ0 1.06 µm
24-5. (a) LC = = = 5893 nm = 8.62 λ0 (b) ∆ n = = = 0.046
4 ∆ n 4 (1.534 − 1.505) 4 LC 4 (5.8 µ m)
24-6. Using Eq. (24 8) and Table 24 2, the half wave voltage is independent of length and found to be
λ0 546 × 10−9 m
VHW = = = 9.84 kV
2r n30 2 (8.56 × 10−12 /V) (1.48)3
24-7. For this cell,

r 3 r V 30.9 × 10−12 426
∆n= n0 E = n30 = (2.29)3 = 7.9 × 10−6
2 2 L 2 .01
2π 2π π
Φ= r n30 V = (30.9 × 10−12 ) (2.29)3 (426) = 1.57 or rad
λ0 632.8 × 10−9 2
Thus, the cell behaves as a quarter wave plate.
24-8. According to Eq. (24 9)

π V
I = Im ax sin 2
2 VHW
λ0 2π 3
Combining VHW = and Φ = r n0 , we have the ratio
2 r n30 λ0
V Φ λ0 2 r n30 Φ
= · =
VHW 2 π r n30 λ0 π
So,

πΦ Φ
I = Imax sin2 = Im ax sin2
2 π 2
(a) Now I = 0 when the argument of sin2 is 0, π, 2 π, ... The first such value above zero makes
π V
Φ/2 = π ⇒ Φ = 2 π ⇒ = π ⇒ V = 2 VHW
2 VHW
That is for Φ = 2 π, 4 π, ..., V = 2 VHW , 4 VHW , ..., respectively.
(b) With an initial HWP, the sin2 curve in Figure (24 6) is simply translated to the left by the angle π.
Thus, at V = 0, I = Im ax , and at V = VHW , I = Imax .
24-9. Both effects are present in media lacking inversion symmetry. The ratio of retardations is
Φp o ck 2 π r n30 V d2 r n30 d2 10 23 (0.01)2
= · = = = 73
Φkerr λ0 2 π K V 2 L K λ0 V L 1 (550 × 10−9) (1 × 104) (0.02)
24-10. The length of the cell is found as,

d d2 (0.015)2
VHW = √ ⇒L= = m = 3.47 m
2KL 2 K VHW2 2 (0.036 × 10−12 ) (30 × 103)2
The required cell is too long to be practical.
24-11. Starting with the given Doppler effect relation,

2 ν up
∆v=
v
λ ν λ
where, where u p = ± vS sin θ and 2 λS sin θ = λ so that u p = ± vS = ± S . Then,
2 λS 2

2ν νS λ λ 1
∆ν = ± = ± ν νS = ± ν νS = ± νS
v 2 v ν
ν ′ − ν = ± νS ⇒ ν ′ = ν ± νS ⇒ ω ′ = ω ± ωS
133

W
24-12. The distance moved by the sound wave is x = vS t = vS .
vlight
1.5 (0.01)

nW
So, x = vS = 3000 = 150nm. The sound wave advances by λ/3.3 for λ = 500 nm.
c 3 × 108
24-13. Adapting the relevant part of Figure 24 11 leads to the diagram below,
∆ kS
(a) From the vector diagram, for small ∆ θ, ∆ θ ≈
k
∆kS
2 π 2 π νS 2 π ∆νS 2 π
k (b) kS = = so ∆ kS = . Also, K = .
∆θ λS vS vS λ
k kS ∆ kS 2 π ∆νS λ λ
θ Then, ∆θ = = = ∆ νS
k vS 2 π vS

∆θ λ D D
(c) The number of resolvable spots is N = = ∆ νS = ∆ ν S = τ ∆ vS
θD vS λ vs
0.01

D
(d) N = ∆ νS = (120 − 80) 106 = 67
vS 5.95 × 103
24-14. The required acoustic frequency is found as, λ = λ0/n = 2 λs sin θ = 2 (vS /νS ) sin θ
2 (1.6) m
νS = (2 n/λ0) vs sin θ = 2500 sin 1◦ = 221 MHz
632.8 × 10−9 m s
24-15. Using Eq. (24 18), λ = (λ0/n) = 2 λS sin θ and from Snell’s law sin θ0 = n sin θ. Therefore,
λ0 λ λ0
λ0
= 2 λS sin θ0/n ⇒ λ0 = 2 λS sin θ0
n
θ θ
kS
632.8 × 19−9 m

λ λ
24-16. Generally, sin θ = = νS = νS = (2.876 × 10−11 s) νS
2 λS 2 vS 2 (11 × 103 m/s)
At νS = 50 MHz: sin θ = 0.001438; θ = 0.0824◦ = 4.94 ′ At νS = 80 MHz: sin θ = 0.002301; θ = 0.1218◦ = 7.91 ′
So, ∆ θ = 2.97 ′
24-17. The rotation of the plane of polarization is β = V B d where for a solenoid B = µ0 n i. For an optical isolator
β = π/4. For Zn S , V = 0.225 min /G cm = 65.45 rad/T m. It is given that the number of turns per meter in
the solenoid is n = 6000/m. The design parameters are then the current i and length d. So,
π π/4 π/4
β = = V B d = V µ0 n i d ⇒ i d = = A · m = 1.59 A · m
4 V µ0 n (65.45) (4 π × 10−7) (6000)
For example for d = 5 cm = 0.05 m, the necessary current is i = 31.8 A.
β 900 min min

24-18. (a) At 632.8 nm, V = = = 0.0647
B d (5098 G) (2.73 cm) G cm
1330 min min
(b) At 543.5 nm, V = = 0.0956
(5098 G) (2.73 cm) G cm
min

24-19. β = V B d = 0.0423 (4000 G) (5 cm) = 846 ′ = 14.1◦
G cm
dn V 0.0423
= = = 0.0712 µm −1
d λ 1.0083 λ 1.0083 (0.589 µ m)
134
Chapter 25 Optical Properties of Materials
25-1. (a) Consider,
K ≡ n2 = (nR + i nI )2 = n2R − n2I + 2 i nI nR = KR + i KI

KR = n2R − n2I , KI = 2 nI nR
Solving these two relations for nR and nI proceeds as,

p
2
KI = 2 nI nR = 2 nI KR + n2I
KI2 = 4 n2I (KR
2
+ n2I

4 n + 4KR nI − KI2 = 0
4 2 2
p I p
2
− 4 KR ± 16 KR + 16 KI2 − KR ± KR 2
+ KI2
n2I = =
8 2
To make n2I > 0, choose the + sign. Thus,
" p #1/2
2
− KR + KR + KI2
nI =
2
p p
2
2 KR − KR + KR + KI2 KR + KR2
+ KI2
n2R = KR + n2I = + =
2 2 2
" p #1/2
2
KR + KR + KI2
nR =
2
(b) If KI ≈ KR,
" p #1/2 √ !1/2
KI + 2 KI2 √ 1+ 2 √
nR = = KI = 1.099 KI
2 2
" p #1/2 √ !1/2

− KI + 2 KI2 √ −1+ 2 √
nI = = KI = 0.455 KI
2 2
25-2. “Nearly transparent”: We assume ω is not too large and use the results of Section 25 5. As explained in the
text between Eqs. (25 29) and (25 30), α = 2 KI . Then using Eq. (25 55),
σµ ω 1/2
0
2 KI = 2
2
Eliminate ω in favor of nR, using Eq. (25 56)
1/2
σ
nR =
2 ω ε0
Then,
σ µ ω 1/2 σ 1/2 σ µ 1/2
0 0
K I nR = =
2 2 ω ε0 2 ε0
1/2
σ µ0 σ
α = 2 KI = = 377
nR ε0 nR
135
25-3. Using Eqs. (25 37) and (25 38) we have,
" # " #
2 2 2
2 2 N e ω 0 − ω N e2 γω
nR − nI ≡ K R = 1 + 2 nI nR ≡ K I =
m ε0 ω02 − ω 2)2 + γ 2 ω 2 m ε0 ω02 − ω 2)2 + γ 2 ω 2
Solving these simultaneously as in problem 25 1 gives,

" p #1/2 " p #1/2
2
KR + KR + KI2 − KR + KR2
+ KI2
nR = nI =
2 2
To compare to Figure 25 2 use the parameters,
ω0 = 1016 s−1 γ = 1014 s−1 N = 1028 m −3
m = 9.109 × 10−31 kg e = 1.6 × 10−19 C ε0 = 8.854 × 10−12 F/m
Using the free plotting program GNUplot:
This is a TeXmacs interface for GNUplot.

GNUplot] w0 1e16~g 1e14~N 1e28~eps0 8.854e-12~e 1.6e-19~m 9.109e-31
KR(x) 1+N*e**2/m/eps0*(w0**2-x**2)/((w0**2-x**2)**2+g**2*x**2)
KI(x) N*e**2/m/eps0*g*x/((w0**2-x**2)**2+g**2*x**2)
nR(x) sqrt((KR(x)/2+sqrt(KR(x)*KR(x)+KI(x)*KI(x))/2))~nI(x) KI(x)/2/nR(x)
set xrange [1e15:1e17]~set samples 1000~set yrange [0.001:10]
set logscale xy~set size 0.6,0.6~set xlabel "{/Symbol w}"
plot nR(x) title "n_R",nI(x) title "n_I"
10
nR
nI
0.1
0.01
0.001
1e+15 1e+16 1e+17
ω
25-4. Given
NA ρ (6.02 × 1026 ) (2.70 × 103) −3
n = 1 electron/atom N = = m = 6.027 × 1028 m −3 σ0 = 3.54 × 107/Ω m
M 26.982
N e2 N e2 (6.027 × 1028 ) (1.602 × 10−19 )2 −1

(a) σ0 = ⇒γ= = s = 4.80 × 1013 s−1
mγ m σ0 (9.109 × 10−31 ) (3.54 × 107)
N e2 (6.027 × 1028 ) e2
(b) ω 2p = = = 1.918 × 1032 s−2 ⇒ ω p = 1.38 × 1016 s−1
m ε0 m ε0
(c) At λ = 550 nm, ω = 2 π f = 2 π c/λ = 3.425 × 1015 Hz. Now,
!
ω2p γ ω 2p
n2R − n2I ≡ KR = 1 − = − 15.347 2 n I n R ≡ K I = = 0.2291
ω2 + γ 2 ω ω2 + γ 2

" p #1/2
2
KR + KR + KI2 K
nR = = 0.0292 nI = I = 3.92
2 2 nR
136
ω2

σ ω µ0 ω
25-5. Before approximation, k = 2 + i 2
. If ω ≪ γ or ≪ 1, we can neglect iω/γ compared with the
c 1 − i ω/γ γ
real part, 1, in the denominator: 1 − i ω/γ . Then
ω2 w2 i σ c2 µ0

1
2
k ≈ 2 + i σω µ0 = 2 1+ . Use c2 ≡
c c ω ε0 µ0
w2

σ
k2 ≈ 1+i .
c2 ε0 ω
σ σ
But if ω ≪ ⇒ ≫ 1, we can neglect the real part, compared with the imaginary, so
ε0, or ε0 ω
w2 σ
k2 ≈ i = i (ω ε µ0) = i ω σ µ0
c2 ε 0 ω
This is the approximation stated at the beginning of Section 25 5 in the text. Equation (25 58) then fol
lows as detailed there.
25-6. Use Eq. (25 58)

r r
1 2 λ
δ= = =
KI σ µ0 ω σ µ0 π c
(a) For ν = 60 Hz ⇒ ω = 2 π (60) s−1

1/2 1/2
2 2
δcu = = m = 0.00856 m = 0.856cm
σ µ0 ω (5.76 × 107) (4 π × 10−7) (2 π) (60)
(b) For λ = 3 m
1/2 1/2
λ 3
δcu = = m = 6.63 µm
σ µ0 π c (5.76 × 107) (4 π × 10−7) π (3 × 108)
1/2 1/2
2 2
25-7. (a) δAl = = m = 0.345mm
σ µ0 ω 3.54 × 107 (4 π × 10−7) 2 π × 6 × 104
3.54 × 107

(b) δs.w . = × δAl = 0.991 m ≈ 1 m
4.3
1/2 1/2
0.1

λ
25-8. δAg = = m = 1.68 × 10−6 m = 1.68 µm
σ µ0 π c 3 × 107(4 π × 10−7) π (3 × 108)
As long as the silver coating is thicker than this the silver plated brass component would work.
25-9. (a) I = I0 e−αx ⇒ (I/I0) = (1/4) = e−αx = e−α(3.42 m ) ⇒ 3.42 α = ln(4) ⇒ α = 0.405 m−1
−1
(b) (I/I0) = (1/100) = e−(0.405 m )x
⇒ (0.405 m−1) = ln(100) ⇒ x = 11.37 m
137
25-10. For a metal we have,
!
ω 2p γ ωp2
n2R − n2I ≡ KR = 1 − 2 n I n R ≡ K I =
ω2 + γ 2 ω ω2 + γ 2
" p #1/2
2
KR + KR + KI2 K
nR = nI = I
2 2 nR
To compare to Figure 25 3 use the parameters,

ω p = 1.63 × 1016 s−1 γ = 4.1 × 013 s−1
Using the free plotting program GNUplot:
This is a TeXmacs interface for GNUplot.
GNUplot] wp 1.63e16~g 4.1e13
KR(x) 1-wp*wp/(x*x+g*g)
KI(x) g/x*wp*wp/(x*x+g*g)
nR(x) sqrt((KR(x)/2+sqrt(KR(x)*KR(x)+KI(x)*KI(x))/2))~nI(x) KI(x)/2/nR(x)
set xrange [1e14:1e18]~set samples 1000~set yrange [0.001:10]
set logscale xy~set size 0.6,0.6~set xlabel "{/Symbol w}"
plot nR(x) title "n_R",nI(x) title "n_I"
10
nR
nI
0.1
0.01
0.001
1e+14 1e+15 1e+16 1e+17 1e+18
ω
25-11. Referring to page 543,

N e2 ω2 ω4 ω2 ω4 α (2 π c)2 α (2 π c)4

n2 ≈ 1 + 2 1 + 2 + 4 ≡ 1 + α 1 + 2 + 4 = (1 + α) + 2 2
+ 4
m ε0 ω0 ω0 ω0 ω0 ω0 ω0 λ ω0 λ4
′ ′
B C
where B = α (2 π c)2/ω02 C = α (2 π c)4/ω02
′ ′ ′ ′
n2 = A + + A =1+α
λ2 λ4
" #1/2 " !#1/2  !2 
B
′
C
′ ′
B /A C /A
′ ′ ′ √ ′ ′
B /2 A
′ ′
C /2 A
′ ′
1 8 /A
′
= A1/2 1 +
′
n= A + + + ≈ A 1 + + − 
λ2 λ4 λ2 λ4 λ2 λ4 8 λ2
( !)
√ ′ ′
1 B /A
′
1 C
′
B2
′
B C
n= A 1+ + 4 ′ − =A+ 2 + 4
2 λ2 λ 2A 8A2
′
λ λ
√ ′
A ≡ A = (1 + α)1/2
′
√ ′ B′ α (2 π c)2

πc
2
1
B ≡ B /2 A = = = 2
2 A 2 ω02 (1 + α)1/2 ω0 (1 + α)1/2
" #
√ ′ C6 ′ ′
B2 (2 π c)4 α (2 π c)4 α2 (2 π c)4 α

α

(2 π c)4 α

4+3α

C≡ A ′ − = − = 1− =C =
4 (1 + α) 2 ω04 (1 + α)1/2 4 (1 + α)
′ ′ ′ ′
2A 8A2 2 ω04 A 1/2 8 ω04 A 3/2 2 ω04 A 1/2
4
πc α (4 + 3 α)
C =2
ω0 (1 + α)3/2
138
25-12. The Lorentz force law is,
F = FE + FB = q E + q v × B
where v is the velocity of the charged particle experiencing the force
|FB | |q v × B | |q v B | |v(E0/c)|
= 6 = = v/c
|FE | |q E | |q E | |E0|
The magnitude of the magnetic force is much less the the magnitude of the electric force so long as the
speed of the charged particle is much less than the speed of light.
139
Chapter 26 Laser Operation
26-1. (a) According to Eq. (26 6), the spectral energy density has the form
8 π h ν3 1
ρ(ν) =
c3 ehν/kBT − 1
The energy per unit volume per unit wavelength interval ρ(λ) is related to ρ(ν) by the relation,
Z λ=∞ Z ν =∞
ρ(λ)d λ = ρ(ν) d ν
λ=0 ν =0
so that integration over all wavelengths or frequencies gives the total energy density. Since, ν = c/λ,
c
dν = − dλ
λ2
So, using Eq. (16 6) and changing variables in the frequency integral proceeds as,
Z ν =∞ Z ν =∞ Z λ=0
8 π h ν3 1 8πh 1 c
ρ(ν) d ν = d ν = − dλ
c3 ehν/kBT − 1 λ=∞ λ
3
ehν/kBT − 1 λ2
Z νν =0
=∞
ν =0
Z λ=∞ Z λ=∞
8πh 1 c 8πhc 1
ρ(ν) d ν = − 3 hν/kBT
− 2
d λ = 5 hν/kBT
dλ
ν =0 λ=0 λ e − 1 λ λ=0 λ e −1
So that
8πhc 1
ρ(λ) =
λ5 ehν/kBT − 1
(b) Generally the time averaged energy density hui of radiation moving in a given direction is related to the
exitance in that direction by the relation M = v hui. (See problem 26 2). If the energy has a variety of
propagation directions the relation becomes
M = (fraction of energy moving in given direction) (average velocity component in that direction) u
The exitance per wavelength energy is the power per area leaving a surface. Let us place the small area in
question so that its surface normal points in the z direction. In black body radiation the energy is dis
tributed isotropically. On average then, half of the “photons” making up the field will have a velocity com
ponent in the z direction. The z component of the velocity vector for a given photon can be calculated as,
vz = v · ẑ = c cos θ
Here θ is the azimuthal angle. The average z velocity component of the photons with a component in the
positive z direction is then found by summing (integrating) the velocity z components associated with the
vectors pointing into a given elemental solid angle dω and then dividing by the total solid angle (2 π) in a
hemisphere,
Z Z 2π Z π/2
1 c c
vz,ave = c cosθdω = dϕ cos θ sin θdθ =
2π 2π 0 0 2
Then the relationship between the energy density and exitance becomes,
1 c
M= hui
22
This relation also holds in each wavelength range so that,
c c 8πhc 1 2π h c2 1
Mλ = ρ(λ) = 5
=
4 4 λ e hν/kB T
−1 λ5 ehν/kBT − 1
140
26-2. Consider the small volume shown below.
Let the total energy in the volume ∆V = ∆A ∆z is W then
∆A W
The energy density in the volume is then hui = .
∆V
In a time ∆t = ∆z/c all of the energy in the volume will pass
through the right face of the volume. So the irradiance I in

∆z W W
can be formed as I = = = c hui.
∆A ∆t ∆A ∆z/c
26-3. Equations (26 14) (in steady state) and (26 16) are,
σI
0 = − A21 N2 − (N2 − N1) (1)
hν ′
NT = N1 + N2 (2)
From (2), N1 = NT − N2. Using this in (1) and rearranging proceeds as,
σI
0 = − A21 N2 − (N2 − NT + N2)
hν ′
N2 A21 + 2 σI/hν ′) = (σI/hν ′) NT
σI/(hν ′) NT σI/(hν ′A21 )
N2 = = NT (3)
(A21 + 2 σI/hν ) 1 + 2 σI/(hν ′A21 )
′
Using (3) in (2) gives,
1 + 2 σI/(hν ′A21 ) σI/(hν ′A21 ) 1 + σI/(hν ′A21 )

N1 = NT − N2 = − NT =
1 + 2 σI/(hν A21 ) 1 + 2 σI/(hν ′A21 )
′ 1 + 2 σI/(hν ′A21 )
Then,
σI/(hν ′A21 ) 1 + σI/(hν ′A21 ) NT
Ninv = N2 − N1 = N T − =−
1 + 2 σI/(hν ′A21 ) 1 + 2 σI/(hν ′A21 ) 1 + 2 σI/(hν ′A21 )
26-4. (a) From Section 26 2,

σNT α0
α = − σ(N2 − N1) = ≡
1 + 2 σI/(hν ′A21 ) 1 + 2 σI/(hν ′A21 )
Then α = α0/2 for
hν ′ hν ′
2 σIS ,abs /(hν ′A21 ) = 1 ⇒ IS ,abs = A21 =
2σ 2στ2
(b) In an ideal four level gain medium the saturation irradiance is given by Eq. (26 39) as
hν ′
IS =
στ2
The difference of the factor of 2 arises because the ideal four level gain medium has a lower lasing level
with a zero lifetime so that it is never populated. Each photon transition from upper to lower level then
changes the population inversion by (N2 − N1) − (N2 − 1 − N1) = 1 since it decreases the population of the
upper level by 1 but does not change the population of the lower level which is always zero. In the absorp
tive medium each absorption changes the inversion by (N2 − N1) − [(N2 + 1) − (N1 − 1)] = − 2 since it
decreases the population of the lower level by one and increases the population of the upper level by 1.
141
26-5. (a) The rate equations are given by Eqs (26 26) (26 29). Setting the indicated parameters to zero:
dN3 σ I
= − κ32 N3 − p p (N3 − N0)
dt hν p
dN2 σI
= κ32 N3 − κ21 N2 − ′ (N2 − N1)
dt hν
dN1 σI
= κ21 N2 + ′ (N2 − N1) − κ10 N1
dt hν
dN0 dN3 dN2 dN1
=− + + ⇒ NT = N0 + N1 + N2 + N3
dt dt dt dt
(b) In steady state and with I = 0 these become,
σ p Ip
0 = − κ32 N3 − (N3 − N0) (1)
hν p
0 = κ32 N3 − κ20 N2 (2)
0 = κ21 N2 − κ10 N1 (3)
NT = N0 + N1 + N2 + N3 (4)
These are four equations in four unknowns. Solving them simultaneously using the free computer algebra
system Maxima I find, (in mks units)
1.499985 × 1031 Ip
Ninv = N2 − N1 = (5)
100000 I p + 1.6565 × 1012
Plotting this expression gives
6e+25 1e+24
9e+23
5e+25 8e+23
4e+25 7e+23
N nv (m-3)
N nv (m-3)
6e+23
3e+25 5e+23
4e+23
2e+25 3e+23
1e+25 2e+23
1e+23
0 0
0 2e+06 4e+06 6e+06 8e+06 1e+07 0 20000 40000 60000 80000 100000
Ip (W/m2) Ip (W/m2)
(c) From the plot or from (5), the inversion is greater than or equal to zero for all pump irradiances.
(d) The small signal gain coefficient is

1.499985 × 1031 Ip
γ0 = σ Ninv = 0.01/cm = 1/m ⇒ Ninv = γ0/σ = 1022 /m3 =
100000 I p + 1.6565 × 1012
Numerically I find Ip = 1104 .4W/m2.
(e) For γ0 = 1/cm = 100/m. Ninv = 1024 /m3, Ip = 1.11 × 105 W/m2.
(f) Solving (1) (4) in steady state for the two pump irradiances, I find,
For I p = 1104 W/m2: For Ip = 1.1 × 105 W/m2:
N0 = 1.49990 × 1026 /m3 N0 = 1.490 × 1026 /m3
N1 = 1.000 × 1017 /m3 N1 = 1.000 × 1019 /m3
N2 = 1.000 × 1022 /m3 N2 = 1.000 × 1024 /m3
N3 = 1.000 × 1017 /m3 N3 = 1.000 × 1019 /m3
It is reasonable to set N0 ≈ NT in both cases.
(g) Using the suggested relations for an ideal 4 level gain medium

κ32 σ pI p γ κ h νp 1 1
γ0 = σR p2 τ2 = σ NT τ 2 ⇒ I p = 0 3 = γ0 1104.3 W/m
κ3 h ν p σ κ32 σ pτ2 NT 1
For γ0 = 0.01/cm = 1/m; I p = 1104.3 W/m2, For γ0 = 1/cm = 1/m; I p = 1.1043 × 105 W/m2
142
26-6. This is conveniently done with the help of the free computer algebra system Maxima:
(%i1) eq1:0 Rp2-k2*N2-Q*(N2-N1)
(%o1) 0 = − (N2 − N1) Q − k2 N2 + Rp2
(%i2) eq2:0 Rp1+k21*N2-k10*N1+Q*(N2-N1)
(%o2) 0 = (N2 − N1) Q + k21 N2 − k10 N1 + Rp1
(%i3) sols:solve([eq1,eq2],[N1,N2])

(Rp2 + Rp1) Q + k21 Rp2 + k2 Rp1 (Rp2 + Rp1) Q + k10 Rp2
(%o3) N1 = , N2 =
( − k21 + k2 + k10) Q + k10 k2 ( − k21 + k2 + k10) Q + k10 k2
(%i4) Ninv:ratsimp(-rhs(sols[1][1])+rhs(sols[1][2]))
(k21 − k10) Rp2 + k2 Rp1

(%o4) −
(k21 − k2 − k10) Q − k10 k2
Here Q = σI/hν ′, so,

(κ21 − κ10 ) R p2 + κ2 R p1 (1 − κ21 /κ10 ) R p2 − (κ2/κ10 ) R p1
N2 − N1 = Ninv = − =
(κ21 − κ2 − κ10 ) (σI/hν ′) − κ10 κ2 κ2 + (1 + κ20 /κ10 ) (σI/hν ′)
26-7.
(a) (b)
dN3 σ I
3 = − κ32 N3 − p p (N3 − N1)
κ3 2 dt hν p
2 dN2 σI
= κ32 N3 − κ21 N2 − ′ (N2 − N1)
σI p/hν p κ2 1 σI/hν ′ dt hν

1 dN1 dN3 dN2
=− + ⇒ NT = N1 + N2 + N3
dt dt dt
(c) In steady state and for I = 0 these relations become,
σ I
0 = κ23 N3 − p p (N3 − N1) (1)
hν p
0 = κ32 N3 − κ21 N2 (2)
NT = N1 + N2 (3)
Solving these, using Maxima, for the given parameters leads to (in mks units),
7.499925 × 1030 I p − 1.242375 × 1038
N2 − N1 = Ninv =
50001 I p + 8.2825 × 1011
1.5e+26 2e+23
1.5e+23
1e+26
1e+23
5e+25 5e+22
Ninv (m-3)
Ninv (m-3)
0 0
-5e+22
-5e+25 -1e+23
-1e+26 -1.5e+23
-2e+23
-1.5e+26
-2.5e+23
-2e+26 -3e+23
0 2e+07 4e+07 6e+07 8e+07 1e+08 1.65e+07 1.655e+07 1.66e+0
Ip (W/m2) Ip (W/m2)
(d) Using the expression for the inversion form part (c), I find Ninv = 0 for Ip = 1.6565 × 107 W/m2.
(e) Setting γ0 = 0.01/cm = 1/m = σNinv = (10 −22 m2) Ninv leads to Ip = 1.6567 × 107 W/m2.
(f) Setting γ0 = 0.01/cm = 100/m = σNinv = (10−22 m2) Ninv leads to Ip = 1.6788 × 107 W/m2.
(g) In a three level system a large pump is needed just to reach a population inversion. Pump thresholds
for lasing are typically not too much more than the pump needed for inversion. In a ideal four level system
an inversion occurs for any nonzero pump.
143
26-8. Consider Eq. (26 41),
I 1
ln L + (IL − I0) = γ0L
I0 IS
If IL ≪ IS and I0 ≪ IS , this relation becomes,

I
ln L ≈ γ0L
I0
IL ≈ I0e γ0L
26-9. Consider Eq. (26 40),

dI γ0
= I
dz 1 + I/IS
If I ≫ IS , this becomes,
dI γ
= 0 I ≈ γ0 IS
dz I/IS
Z IL Z L
dI = γ0 IS dz
I0 0
IL − I0 = γ0 IS L
IL = I0 + γ0 IS L
26-10. (a) Using Eq. (26 38) and (26 39) for ideal 4 level gain media in the expression from problem 26 9 gives,
′
hν
IL − I0 = γ0 IS L = σ R p2 τ2 L = hν ′R p2 L
στ2
(b) The number of photons added to the field per time due to a gain medium of cross sectional area A and
length L is,
Noutput N
= (IL − I0) A/hν ′ = R p2 A L = R p2 V = pum p
t t
Here R p2 is the pump rate per volume and V = A L is the volume of the gain medium. Thus in the large
input irradiance limit every pump event leads to an output photon.
(c) For lower input irradiance some pump events lead to nonradiative decay rather than stimulated emis
sion. (Of course for non ideal gain media there are many other processes that reduce the pump efficiency in
both limits.)
26-11. The relation between the input and output irradiance is,

I 1
ln L + (IL − I0) = γ0 L
I0 IS
For the parameters given in the problem:

I 1
ln L + (IL − I0) = 2
I0 100 W/cm2
One must solve the relation numerically. I find,
(a) When I0 = 1 W/cm2, IL = 6.96 W/cm2. (b) When I0 = 10 W/cm2, IL = 49.7 W/cm2
(c) When I0 = 100 W/cm2, IL = 221W/cm2 (d) When I0 = 1000 W/cm2, IL = 1180 W/cm2
(e) When I0 = 10, 000 W/cm2, IL = 10, 200 W/cm2
144
26-12. Using the results from problem 26 11:
(a) IL − I0 = 5.96 W/cm2, IL/I0 = 6.96 (b) IL − I0 = 39.7 W/cm2, IL/I0 = 4.97
(c) IL − I0 = 121 W/cm2, IL/I0 = 2.21 (d) IL − I0 = 180 W/cm2, IL/I0 = 1.18
(e) IL − I0 = 200 W/cm2, IL/I0 = 1.02
Note that as the input irradiance is increased, the added irradiance becomes a smaller and smaller fraction
of the input irradiance. This is what is meant by gain saturation.
26-13. The relation between the input and output irradiance is,

I 1
ln L + (IL − I0) = γ0 L
I0 IS
When detuned by ∆ν/2, the lineshape function is 1/2 its peak value. The small signal gain coefficient is
proportional to g(ν ′) and the saturation irradiance is inversely proportional to g(ν ′). So for the case at
hand γ0(ν ′) = γ0(ν0)/2 = 1/cm and IS (ν ′) = 2IS (ν0) = 200 W/cm2. For these parameters

I 1
ln L + (IL − I0) = 1
I0 200 W/cm2
One must solve the relation numerically. I find:
(a) When I0 = 1 W/cm2, IL = 2.70 W/cm2, IL − I0 = 1.70 W/cm2, IL/I0 = 2.70
(b) When I0 = 10 W/cm2, IL = 25.2 W/cm2, IL − I0 = 15.2 W/cm2, IL/I0 = 2.520
(c) When I0 = 100 W/cm2, IL = 181W/cm2, IL − I0 = 81 W/cm2, IL/I0 = 1.81
(d) When I0 = 1000 W/cm2, IL = 1170 W/cm2, IL − I0 = 170 W/cm2, IL/I0 = 1.17
(e) When I0 = 10, 000 W/cm2, IL = 10, 200 W/cm2, IL − I0 = 200 W/cm2, IL/I0 = 1.02
26-14. The plot is shown below. I used Maxima to numerically solve Eq. (26 41) for the given parameters and
then plotted the generated list of points. In the plot the top (dashed) curve is for γ0 = 2/cm and IS =
10 W/cm2, the middle (solid) curve is for γ0 = 1/cm and IS = 20 W/cm2. The bottom (dotted) curve is for
γ0 = 1/cm and IS = 10 W/cm2.
160
140
120
100
IL (W/cm )
2
80
60
40
20
0
0 1 2 3 4 5 6 7 8 9 10
L (cm)
145
26-15. (a) Use Eq. (26 45) with S = R3 = 1 − T3. Then

(γ0 − γth ) L (γ0 − γth ) L (γ0 − γth ) L
Iout = T3 IS = T3 IS = T3 IS = IS (γ0 − γth ) L
1−S 1 − (1 − T3) T3
(b) If γ0 ≫ γth ,
Iout ≈ IS γ0 L
As in problem 26 10, using Eq. (26 38) and (26 39) for ideal 4 level gain media in the expression above
′
hν
Iout = γ0 IS L = σ R p2 τ2 L = hν ′R p2 L
στ2
The number of photons emitted into the output field per time due to a gain medium of cross sectional area
A and length L is then,
Noutput N
= (IL − I0) A/hν ′ = R p2 A L = R p2 V = pum p
t t
Here R p2 is the pump rate per volume and V = A L is the volume of the gain medium. Thus, in the large
input irradiance limit every pump event leads to an output photon.
(c) The output efficiency is the ratio of output power to pump power not the ratio of the rate of photons
leaving the cavity to the pump rate. That is the efficiency for the case of part (b) would be,
Noutput /t) h ν ′ ν ′
efficiency = = <1
Npum p /t) hν p νp
1
26-16. (a) γth = ln(1/R3) = 0.00513/cm
L
(γ − γth ) L (2 γth − γth ) L 0.00513 × 10
(b) Iout = T3 IS 0 = T3 IS = (0.05) (2000) W/cm2 = 102.6 W/cm2
1−S 1 − R3 0.05
26-17. The output irradiance has the form,

(γ0 − γth ) L
Iout = T3 IS
1−S
It is given that R3 = 0.95 = 1 − T3 so A3 = 0. Further R1 = R2 ≡ R. Then
S = R1 R2 R3 = R2 (0.95); γth L = ln(1/S)
So,
(0.01) (10) − ln(1/(0.95R2)) 2 0.1 − ln(1/(0.95R ))
2
Iout = (0.05) (2000 W/cm2 100 W/cm

=
1 − 0.95 R2 1 − 0.95 R2
Plotting gives,
100
90
80
70
Iout (W/cm2)
60
50
40
30
20
10
0
0.98 0.985 0.99 0.995 1
R
146
26-18. The output irradiance has the form,

(γ0 − γth ) L
Iout = T3 IS
1−S
From the given information, R3 = 1 − T3 − 0.01 = 0.99 − T3, R1 = R2 = 0.99, γ0 = 0.01/cm, IS = 2000 W/cm2
and L = 10 cm. Then S = 0.99)2 (0.99 − T3)
0.1 + ln((0.99)2 (0.99 − T3))

Iout = 2000 W/cm T32

1 − 0.99)2 (0.99 − T3)
The plot looks like,
45 40.8631
40 40.863
35 40.863
40.8629
Iout (W/cm2)
Iout (W/cm2)
30
40.8629
25
40.8628
20
40.8628
15 40.8627
10 40.8627
5 40.8626
0 40.8626
0 0.01 0.02 0.03 0.04 0.05 0.0241 0.02415 0.0242 0.02425 0.0243
T3 T3
(b) From the plot on the right it is evident that the optimum T3 is just a bit more that 0.0242. Such an
optimum value exists whenever there are losses other than output losses in the cavity. If the transmission
coefficient is too small the light generated in the cavity primarily ends up wasted into the other loss mecha
nisms. If the transmission coefficient is too large, the net gain per pass is reduced and less of the energy
stored in the population inversion is extracted by stimulated emission.
26-19. Consider the linear cavity shown below,
Saturation is now due to both the right and left going

fields in the cavity. Equation (26 40) can be rewritten
I1+ Gain Cell I2+ Iout as,
1 dI+ γ0 1 dI−
I1− I2− = =−
I+ dz 1 + (I+ + I−)/IS I− dz
R1 R2, T2
L The minus sign is due to the fact the the left going
wave propagates in the − z direction.
This relation leads to,
1 dI+ 1 dI − d
+− = 0 ⇒ [ln(I+ I−)] ⇒ I+ I− = C
I+ dz I− dz dz
where C is a constant with dimensions of squared irradiance. Using this fact, one can write,
1 dI+ γ0 1
= ⇒ (1 + (I+ + C/I+)/IS ) dI+ = γ0dz
I+ dz 1 + (I+ + C/I+)/IS I+
Z I2+ Z L
1 1 C I 1 C 1 1
+ + 2 dI+ = γ0 dz ⇒ ln 2+ + (I2+ − I1+) − − = γ0 L
I1+ I+ IS I+ IS 0 I1+ IS IS I2+ I1+
Now the relation between I2+ and I1+ can be found by examining the irradiance at the ends of the gain
cell and including the effects of the mirrors:
√ 2
I1+ I1− = I2+ I2− = C ⇒ I1+ ( I1+/R1) = I2+ (I2+R2) ⇒ I1+ = I2+ R2R1 and C = I2+ R2
Using these in the relations above gives,
√ 2

1 1 I2+ I2+ R2 1 1
ln + 1 − R1 R2 − −√ = γ0 L
2 R1 R2 IS IS I2+ R2 R1 I2+
1
γ0 L − 2 ln(1/(R1 R2) T I γ0 (2L) − ln(1/(R1 R2)
Iout = T2 I2+ = T2 IS √ = 2 S
2 1 − √R R 1 + pR /R
p
1 − R1 R2 − R2 + R2/R1 1 2 2 1
147
26-20. When R1 = 1, Eq. (26 47) becomes,
T2 IS γ0(2 L) − ln(1/R2) T I γ (2 L) − ln(1/R2) T2 IS γ0(2 L) − ln(1/S)
Iout = √ = 2 S 0 =
2 (1 − R2) (1 + R2 ) 2 1 − R2 2 1−S
p
This differs from the ring cavity result given in Eq. (26 43) by the presence of two different factors of 2.
The linear cavity result has γ0(2 L) instead of γ0 L because the field encounters the gain medium twice in
each round trip through the cavity. The linear cavity result has IS /2 rather than IS because in the linear
cavity the gain medium is saturated by both the right and left going fields but the output is proportional
only to the right going field.
26-21. a)At line center,

A21 c3 2 h ν0 A21 c2 1 A21 λ20 (0.34) 10.6 × 10 −6)2

σ = B21 g(ν0) h ν0/c = = = =
8 π h ν03 π∆νH c 4 π 2ν02 ∆νH 4 π 2∆νH 4 π 2 (109)
σ = 9.69 × 10−22 m2
(b) Ninv = γ/σ = (3/m)/(9.69 × 10 −22 m2)=3.1 × 1021 /m3
h ν0 6.626 × 10−34 ) (3 × 108/10.6 × 10−6)
(c) IS ≈ = = 1.94 × 106 W/m2
σ τ2 (9.69 × 10 −22 ) (10 × 10 −6)
26-22. The Doppler width is

1/2 1/2
8 · 1.38 × 10−23 · 400 3 × 108

8 kB T
∆νD = ln(2) ν0 = ln(2) = 1.5 × 109 Hz
M c2 (20.2 · 1.66 × 10−27 ) 9 × 1016 6.33 × 10−7
26-23. Consider the ring cavity shown in Figure (26 8) in the text. Let the field just before M3 in that figure at
time t + τ , where τ = P /c is the cavity round trip time be
E(t + τ ) = E0(t + τ ) eiω(t+τ )
Without a gain medium or a field injected into the cavity, this field is formed by the propagation of the
field that exists just before mirror three at time t around the cavity. That is,
√
E0(t + τ ) eiω(t+τ ) = S ei(ωτ −kP )E0(t)eiωt
√ −ikP
E0(t + τ ) = S e E0(t)
If the field is resonant with the cavity, e−ikP = 1. So, with a Taylor series expansion of the left hand side,
dE0(t) √
E0(t) + τ = S E0(t)
dt
Rearranging, integrating, and forming the irradiance gives,
√
dE0(t) (1 − S )
=− E0(t) ≡ − (Γ/2)E0(t)
dt τ
E0(t) = E0(0) e−Γt/2
I ∝ E02 ⇒ I = I0e−Γt
Here √ √
1− S 1− S
Γ=2 =2 c
τ P
Now the expansion leading to this result is only good for S close to 1. So let S = 1 − L where L ≪ 1. Then
√ √
1 − S = 1 − 1 − L ≈ 1 − (1 − L/2) = L/2 = (1 − S)/2
So, √
1− S 1−S
Γ=2 c≈ c
P P
148
26-24. Figure 26 13 was made by using the free computer algebra system Maxima to numerically integrate Eqs.
(26 54) and (26 55). The script I used is shown below.
(%i1) (sc:3e-8,d:1e-9,Vr:0.3,G:1e8,A:1e7,Rth:G*A/sc/Vr,y:2)$
(%i2) R:y*Rth$
(%i3) (Np[1]:0.0,N2[1]:0.0)$
(%i4) for j:2 thru 1500 do
(N2[j]:N2[j-1]+d*(R-A*N2[j-1]-sc*Np[j-1]*N2[j-1]),
Np[j]:Np[j-1]+d*(sc*Vr*(Np[j-1]+1)*N2[j-1]-G*Np[j-1]))$
(%i5) (N2list:listarray(N2),Nplist:listarray(Np),t:makelist(d*(k-1)/1e-6,k,1,1500))$
(%i6) plot2d([[discrete,t,N2list],[discrete,t,20*Nplist]],[gnuplot_curve_styles, ["with lines 1", "with lines
1"]],[gnuplot_preamble, "set nokey;set samples 2000;set xlabel ’Time ({/Symbol m}s)’;set ylabel ’Number
density (cm^{-3})’;set xrange [0:1.5];set label 1 ’{/Italic N}_{inv}’ at 1.2,1.2e16;set label 2 ’{/Italic
N_p}’ at 1.2,0.73e16 "])$
(%i8) plot2d([[discrete,t,N2list],[discrete,t,20*Nplist]],[gnuplot_term, ps], [gnuplot_out_file,
"fig26_13.eps"],[gnuplot_curve_styles, ["with lines 1", "with lines 2"]],[gnuplot_ps_term_command, "set
term postscript eps enhanced monochrome 14"],[gnuplot_preamble, "set nokey;set samples 2000;set xlabel
’Time ({/Symbol m}s)’;set ylabel ’Number density (cm^{-3})’;set xrange [0:1.5];set label 1 ’{/Italic
N}_{inv}’ at 0.73,1.2e16;set label 2 ’20{/Italic N_p}’ at 0.78,0.8e16;set size 0.7,0.7"]);
26-25. In figure (a), (b), and (d) below the solid curve represents Ninv and the dotted curve is 20 N p. In figure (c)
the solid curve is Ninv and the dotted curve is 100 N p.
7e+16 1.8e+16
1.6e+16
6e+16
1.4e+16
5e+16
Number dens ty (cm-3)
1.2e+16
4e+16 1e+16
3e+16 8e+15
6e+15
2e+16
4e+15
1e+16
2e+15
0 0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.5 1 1.5 2 2.5 3 3.5
(a) Time (µs) (b) Time (µs)
1.4e+16 1.2e+17
1.2e+16
1e+17
1e+16
Number dens ty (cm )
8e+16
-3
8e+15
6e+16
6e+15
4e+16
4e+15
2e+16
2e+15
0 0
0 1 2 3 4 5 6 0 0.2 0.4 0.6 0.8 1 1.2 1.4
(c) Time (µs) (d) Time (µs)
26-26. (a) The fields constructively interfere when they differ in phase by an integer multiple of 2 π. This occurs
for t = n/δν where n is an integer.
(b) In a mode locked laser adjacent cavity modes are phase locked and differ in frequency by the inverse of
the round trip cavity time. Thus these waves will constructively interfere once every round trip.
149
26-27. The pulse width is roughly the inverse of the gain bandwidth ∆ν = ν ∆λ/λ = c ∆λ/λ2.
1 λ2 4882
Ar+: ∆t p ≈ = = s = 1.98 × 10−10 s
∆ν c ∆λ 3 × 1017 · 0.004
1 λ2 6332
He Ne ∆t p ≈ = = s = 6.68 × 10−10 s
∆ν c ∆λ 3 × 1017 · 0.002
1 λ2 5902
Dye ∆t p ≈ = = s = 1.45 × 10 −14 s
∆ν c ∆λ 3 × 1017 · 80
26-28. The plots for the three cases are shown below. I’ll take the pulse width is to be approximately the full
width at half maximum of these waveforms.
6 15 60
50
4 10 40
30
2 5 20
10
0 0
F
F
0
-2 -5 -10
-20
-4 -10 -30
-40
-6 -15 -50
-1.5 -1 -0.5 0 0.5 1 1.5 -1 -0.5 0 0.5 1 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4
t (s) t (s) t (s)
(a) (b) (c)
(a) ∆ν ∼ (10 + 5 (0.2) − 10)Hz = 1 Hz, ∆t p ∼ 1 s, ∆ν ∼ 1/∆t p.
(b) ∆ν ∼ (10 + 10 (0.2) − 10)Hz = 2 Hz, ∆t p ∼ 0.5, ∆ν ∼ 1/∆tp.
(c) ∆ν ∼ (10 + 50 (0.2) − 10)Hz = 10 Hz, ∆t p ∼ 0.1, ∆ν ∼ 1/∆t p.
26-29. Let E = Re(Ẽ ), where,

(N −1)/2
X
Ẽ = E0 ei(2πν0 t+ ϕ0) ei(2πjνf s r t
j =−(N −1)/2
For this formula to be sensible N must be odd. Now let x = ei(2πνf s r t). Then,
 
(N −1)/2
X x(N −1)/2+1 − x−(N −1)/2
Ẽ = E0 ei(2πν0 t+ ϕ0) x j = E0 ei(2πν0 t+ ϕ0)

x−1
j =−(N −1)/2
ei(Nπνf s r t)eiπνf s t t − e−iNπνf s r teiπνf s t t i(2πν0 t+ ϕ0) e

iπνf s t t i(Nπνf s r t)
e − e−iNπνf s r t
Ẽ = E0 ei(2πν0 t+ ϕ0) = E 0 e
ei2πνf s r t − 1 eiπνf s t t eiπνf s t t − e−iπνf s t t
sin(N π νfsr t)
Ẽ = E0 ei(2πν0 t+ ϕ0)
sin(π νfsr t)
sin(N π νfsr t)
E = Re Ẽ = E0 cos(2 πν0 t + ϕ0)
sin(π νfsr t)
150
26-30. The power for the electric field of problem 26 29 is given by P ∝ E 2 where the brackets indicate a time

average over a period long compared to an optical period but short compared to a round trip cavity time.
So,
sin2(N π νfsr t) sin2(N π νfsr t)
P ∝ E02 ⇒ P = P0
sin (π νfsr t)
2 sin2(π νfsr t)
This function peaks when its denominator tends to zero. At these times t = n/νfsr where n is an integer. At
these times the numerator also tends to zero. L’Hopital’s rule gives,
sin(N α) N cos(N α)
Lim = Lim =±N
α→nπ sin(α) α→nπ cos (α)
so that,
sin2(N α)
Lim = N2
α→nπ sin2(α)
and thus,
P p = P0 N 2
The power is periodic in time and so the width of the pulse near t = 0 is the same as the width of all of the
pulses. The first minimum zero occurs when the denominator goes to zero for finite t. This occurs when
1
N π νfsr t1 = π ⇒ t1 =
Nνfsr
The width of the pulse from zero to zero on either side of the central maximum is 2 t1 and so ∆t p ∼ t1 is a
reasonable measure of the pulse width. Thus,
1 1
∆t p ∼ =
N νfsr ∆ν
where ∆ν = N νfsr is the range of frequencies that make up the signal.
26-31. The pulse repetition rate PRR is once per cavity round trip:
c 3 × 108
PRR = 1/(2 d/c) = = Hz = 108 Hz
2d 2(1.5)
The peak power is
Pave 10 W
Pp = = = 1430 W
PRR ∆t p (108)(70 × 10 −12 )
26-32. Using the rough relation for the full-width angular spread due to diffraction from a slit, ∆θ = 2 λ/b where
b is the slit width,
2 λ 2 × 8 × 10 −7
“vertical spread” ∆θ ≈ = = 1.6 rad ≈ 90◦
b 10 −6
2 λ 2 × 8 × 10 −7
“horizontal spread” ∆θ ≈ = = 0.16 rad ≈ 9◦
b 10 × 10−6
26-33. The band gap energy is the energy of the emitted photon,
h c 1240 eV · nm
E= = = 1.55 eV
λ 800
26-34. Assuming equal reflectance R from each cleaved end leads to the threshold condition
R 2 e 2 γ0 L = 1
Solving for the R,

R = e− γ0 L = e−(40)(0.02) = 0.45
151
26-35. (a) In steady state Eqs. (26 54) and (26 55) become,
0 = R p2 − κ2 N2 − σc N p N2 (1)
0 = − Γ N p + (V g/Vc) σ c N p N2 (2)
From (2),
Γ Vc
N2 =
σc V g
Using this in (1) gives,
κ2 Γ Vc
R − κ2 N2 R p − σc Vg σc (V g/Vc) R p2 − κ2 Γ
Np = p2 = =
σc N2 Γ Vc/V g σcΓ
c
(b) In problem 26 23 it is shown that Γ = (1 − S). Also V g/Vc = L/P where I have taken the cross sec
P
tional area A of the beam to be the same as that of the gain medium. The rate that photons leave the
cavity is given by
d(N p Vc) c 1 d(N p Vc) 1

= T3 VcNp ⇒ Iout = h ν ′ = h ν ′ T3 (c/P ) VcN p = T3h ν ′ c N p
dt P A dt A
σc (L/P ) R p2 − κ2 (1 − S) c/P κ hν ′ (σR p2 L/κ2) − (1 − S)

Iout = hν ′ c T3 = T3 2
σc (1 − S) c/P σ 1−S
(c) Now, κ2 = 1/τ2 so that

κ2 hν ′ hν ′ σR p2
= = IS and = σR p2 τ2 = γ0
σ στ2 κ2
Then,
γ0 L − (1 − S)
Iout = T3 IS
1−S
This agrees with Eq. (26 43) only for parameters such that
1 − S ≈ ln(1/S)
This is approximately valid if S = 1 − L is close to one so that L ≪ 1. Then

1
ln(1/S) = ln = − ln(1 − L) ≈ − ( − L) ≈ L = 1 − S (L ≪ 1)
1−L
So for 1 − S ≪ 1, (low losses) the approach taken in this problem agrees with the approach taken in the
body of the text leading to Eq. (26 43).
(d) If S is not close to one the approach leading to Eq. (26 43) remains valid. The formation of the rate
equation for Np given as Eq. (26 55) is not valid for large losses as the approach borrowed the cavity loss
term from the approach taken in Chapter 8 (see also the solution to problem 26 23) which requires that the
loss per round trip be small for its validity.
152
Chapter 27 Characteristics of Laser Beams
27-1. (a) The TEM00 Gaussian beam is similar to a plane wave in that the direction of energy flow is primarily
in a given direction and at the beam waist and in the far field the phase fronts are planar. The irradiance
in a TEM00 beam dies off in the radial direction whereas a plane wave represents the (impossible) situation
for which the wave exists in all space.
(b) TEM00 Gaussian beams have nearly spherical phase fronts but the center of curvatures of these phase
fronts changes with distance from the beam waist. The center of curvature of all of the phase fronts of a
spherical wave is at the position of the source of the wave. A TEM 00 beam is confined in the (cylindrical)
radial direction and propagates primarily in a given linear direction. Spherical waves propagate radially
outward from the source.
27-2. Eq. (27 3) has the form,

n2 ∂ 2Ẽ
∇2Ẽ − =0
2 c2∂t2
∂ ∂2 ∂2 k 2 ∂ 2Ẽ
2
+ 2 + 2 Ẽ − 2 2 = 0
∂x ∂y ∂z ω ∂t
Take,
Ẽ = U (x, y, z) ei(kz −ωt+ ϕ)
The necessary x, y and t derivatives are easy to obtain.
∂2 ∂2
2
Ẽ = ei(kz −ωt+ϕ) 2 U
∂x ∂x
∂2 ∂2
2
Ẽ = ei(kz −ωt+ϕ) 2 U
∂y ∂y
∂ 2Ẽ
= ei(kz −ωt+ ϕ) − ω 2 U

∂t2
For z and t a bit more work is required.
∂Ẽ ∂U
= ei(kz −ωt+ ϕ) + i k U ei(kz −ωt+ ϕ)
∂z ∂z
∂ 2Ẽ

∂ i(kz −ωt+ ϕ) ∂U i(kz −ωt+ ϕ)
= e + i k U e
∂z 2 ∂z ∂z
∂ 2Ẽ i(kz −ωt+ ϕ) ∂U

2
i(kz −ωt+ϕ) ∂ U ∂U
= i k e + e + i k ei(kz −ωt+ ϕ) − k 2 U ei(kz −ωt+ ϕ)
∂z 2 ∂z ∂z 2 ∂z
∂ 2Ẽ ∂ 2U

i(kz −ωt+ ϕ) i(kz −ωt+ ϕ) ∂U 2 i(kz −ωt+ ϕ)
=e 2ike −k Ue + 2
∂z 2 ∂z ∂z
Similarly,
∂ 2Ẽ ∂ 2U

i(kz −ωt+ ϕ) i(kz −ωt+ ϕ) ∂U 2 i(kz −ωt+ ϕ)
= e − 2 i ω e − ω U e +
∂t2 ∂t ∂t2
Using these all in the wave equation gives
2
∂ U ∂ 2U ∂ 2U k2 2

∂U
ei(kz −ωt+ ϕ) + + + 2 i k − k 2
− ω U =0
∂x2 ∂y2 ∂z 2 ∂z ω2
∂ 2U ∂ 2U ∂ 2U ∂U
+ 2 + 2 +2ik =0
∂x2 ∂y ∂z ∂z
153
27-3. We wish to integrate,
dp i
=
dz z − i z0
Separation of variables and subsequent integration proceeds as,

Z p(z) Z z
1
−i dp = ′
dz ′
p(0) 0 z − i z0

z − i z0
− i (p(z) − p(0)) = ln(z ′
− i z0)|z0 = ln = ln(1 + i z/z0)
− i z0
s
p −1 z 2 + z02 i tan −1(z/z0)
e−ip(z) eip(0) = 1 + i z/z0 = 1 + z 2/z02 ei tan (z/z0) = e
z02
s
z02 −itan −1(z/z0) ip(0)
eip(z) = 2 2e e
z + z0
The phase factor eip(0) can be absorbed into the overall phase factor in the waveform.
27-4. (a) The beam waist occurs at z = 0. So, using Eqs. (27 14) and (27 21)
π w02 π (5 × 10 −4)2
q(0) = − i z0 = − i =−i m = − i (1.24 m)
λ 632.8 × 10−9
(b) Using the result from (a),
q(z) = z − i z0 ⇒ q(50 m) = 50 m − i (1.24 m)
It would be a waste of effort to use

1 1 λ
= +i
q R π w 2(z)
directly since it would simply lead algebraically to the q = z − i z0 through the use of Eqs (21 29) and (27
20). However it is instructive to note that since z = 50 m ≫ z0 = 1.24 m and so R ≈ z. Further the spot size
in the far field is given approximately by,
π
= z θFF = 50 × 4 × 10−4 m = 0.02 m

w(z) = z
λ w0
1 1 λ 1 632.8 × 10−9 1
= +i ≈ + i ⇒q≈ m = 49.97 m − i (1.26 m)
q R π w (z) 50 m
2 π (0.02) m
2 0.02 + i(5.036 × 10 −4)
This gives a feel for the nature and appropriateness of the far field approximations.
27-5. (a), (b) Using Eqs. (27 19) and (27 20) together with some of the results from the solution to problem 27 4,
" 2 #
z02 1.24

R(z) = z 1 + 2 = (50 m) 1 + = 50.03 m ≈ z
z 50
r
λz0
1 + z 2/z02 = 0.02016 ≈ z θFF = 0.02 m

w(z) =
π
154
27-6. (a) From symmetry considerations the beam waist must be at the center of the cavity. More formally,
using Eqs (27 32) and (27 33) as follows,
RM 2 z 1 + z02/z22 z2/z0 1 + z02/z22 u2 1 + 1/u22

=−1= 2 = ≡ u1,2 ≡ z1,2/z0
RM 1 z1 1 + z02/z22 z1/z0 1 + z02/z22 u1 1 + 1/u21
u 1 + u22
−1= 1 ⇒ u22 u1 + u2(1 + u21) + u1
u2 1 + u21
− 1 − u21 ± (1 + u21)2 − 4 u21 − 1 − u21 ± 1 − u21)
p
u2 = = = − u1 or − 1/u1
2 u1 2 u1
For u2 = − u1, z2 = − z1. Then using Eq. (27 34), z2 − z1 = 2 z2 = d ⇒ z2 = d/2, z1 = − d/2. This is the solu
tion of interest.
But what about the other solution? For this case z1 = − z02/z2 and Eq. (27 34) gives
z2 + z02/z2 = z2(1 + z02/z22 = d

But z2(1 + z02/z22 = RM2 and since in this case RM 2 does not equal d, the second solution is not valid for

this case.
All in all, it is best to use the symmetry argument when it is available.
s s
10

RM 1
(b) Use, R(z) = RM2 = z2 1 + z02/z22 2
− 1 (0.17 m),

⇒ z0 = − 1 z1 =
z1 0.17
p
z0 = 1.293 m = π w02/λ ⇒ w0 =
p
z0 λ/π = (1.293) (632.8 × 10−9)/π = 5.10 × 10−4 m = 0.510 mm
(c) The spot sizes at the mirrors are equal to each other and each have the value,
p p
w(z2) = w0 1 + z22/z02 = (0.510 mm) 1 + 0.172/1.2932 = 0.514 mm
λ 632.8 × 10−9
(d) θFF = = = 0.395 mrad
π w0 π (5.10 × 10 −4)
(e) zFF ≫ 50 (1.293 m) = 64.6 m
(f) The irradiance at the center of the beam waist is found (see problem 27 13) from,
2 Φtot (0.01 W)
I0 = = = 1.22 W/cm2
π w02 π (0.051 cm)2
Using Eq. (27 26),
z02 1.2932

= 1.22 W/cm2
= 4.89 × 10−4 W/cm2

I(ρ = 0, z = z FF ) = I0
(zFF )2 + z02 64.62 + 1.2932
RM 1 RM 2
w(z2)
w0 θFF
z1 = − d/2 z2 = d/2
z=0
155
27-7. (a) The ABCD matrix is,
         
1 0 1 t 1 0 1 0 1 0.004 1 0 1.0007 0.0027
         
=
1 =
       
 n − n′ n   n − n ′ n   0.5 1.5   − 0.5 
0 1 0 1 − 0.5318 0.9979
R3n ′ n′ R2 n ′ n′ − 0.64 1 − 1.5( − 2) 1.5
(b) For t → 0:
     
1 0 1 0 1 0 1 0
     
= =
   
0.5 1.5   − 0.5 1   
0 1 − 0.53125 1
− 0.64 1 − 1.5( − 2) 1.5
R3
R2
n
n′ n′
27-8. (a) From the overall matrix in the solution to problem 27 7b,
1
C = − 0.53125 = − ⇒ f = 1.88 m
f
(b) Using the lensmaker’s formula for a thin lens,
1.5 − 1

1 n2 − n1 1 1 1 1
= − = − = 1.88 m
f n1 R1 R2 1 − 2 m − 0.64 m
Figure for Problems 27-9 and 27-10
d ℓ
t
R1 = ∞
R2 R3
q2
q1
156
27-9. (a) The plane mirror has R = ∞ so that.
1 1 λ 1 λ π w12
= +i = + i ⇒ q1 = − i = − i z01
q1 R1 π w12 ∞ π w12 λ
Note that what is denoted as w1 here is the spot size at a the beam waist of the beam in the cavity.
(b) Using Eq. (27 19), with z = d = 0.7 m, R = R(z) = 2 m

r r
2 R 2
/z 2 ⇒ z0 = z − 1 = (0.7 m) − 1 = 0.954 m

R2 = z 1 + z01
z 0.7
r
632.8 × 10−9 (0.954)
r
π w12 λ z01
z01 = ⇒ w1 = = = 4.38 × 10−4 m = 0.438 mm
λ π π
(c) Combining the results of parts (a) and (b),
q1 = − i z01 = − i (0.954 m)
(d) At the beam waist outside the cavity,
1 C q1 + D A q1∗ + B A C q1 q1∗ + B C q1 + A D q1∗ + B D
= =
q2 A q1 + B A q1∗ + B A2 q1 q1∗ + B 2
Here, in mks units, q1 = − i (0.954) and,
A = 1 − 0.53 ℓ B = 0.7 + 0.63 ℓ
C = − 0.53 D = 0.63
(e) Using Eq. (27 17), R2 = ∞ and the result from (d),
1 1 λ 1 1 A C q1 q1∗ + B C q1 + A D q1∗ + B D
= +i = +i =
q2 R2 2
π w2 R2 z02 A2 q1 q1∗ + B 2
The values for the various parameters listed in part (d). Equating real parts, (in mks units)
A C q1 q1∗ + B D
0= ⇒ A C q1 q1∗ = − B D
A2 q1 q1∗ + B 2
(1 − 0.53 ℓ) ( − 0.53)(0.954)2 = − (0.7 + 0.63 ℓ 0.63) ⇒ ℓ = 0.063 m = 6.3 cm

With this, A = 0.967, B = 0.740, C = − 0.53, and D = 0.63. Equating imaginary parts,
r
π w22 A2 q1 q1∗ + B 2 λ z02
z02 = = = 1.464 m ⇒ w2 = = 5.4 × 10−4 m = 0.54 mm
λ ( − B C + A D) |q1| π
27-10. (a) If one treats the output mirror/lens as a thin lens Eqs. (27 46) and (27 47) can be used to predict the
location and size of the beam waist. From the solution to problem 27 8a, f = 1.88 m. Then using some of
the results of the previous series of problems,
2 2 !2
0.7 π 4.38 × 10−4

1 1 Z1 1 π w01 2 1 1
2 = 2 1− + 2 = 1−
1.88
+
632.8 × 10−9
w02 w01 f f λ (4.38 × 10−4 m)2 (1.88 m)2
w02 = 5.43 × 10−4 m = 0.543 mm
2
f 2 (Z1 − f ) (1.88) (0.7 − 1.88)
and Z2 = f + 2 = 1.88 m + m = 0.069 m ≈ 7 cm
(Z1 − f ) + z01 2
(0.7 − 1.88)2 + 0.9542
These answers correspond reasonably well to the answers found with the more exact and exacting for
malism of the previous solution.
2 2
(b) These approximate formulas are good only if (see the discussion following Eq. (27 48)) z01 ≫ (Z1 − f ) .
For the case at hand,
2
z01 = (0.954 m)2 = 0.910 m2, (Z1 − f )2 = (0.7 − 1.88)2 = 1.3924 m2
So the validity criterion is not met.
157
27-11. (a) The far field distance for the external beam is
zFF = 50 z02 = 50 (1.464 m = 73.2 m

Note that z02 was calculated in the solution to problem 27 9e.
λ 632.8 × 10 −9
(b) The far field divergence angle is, θFF = = = 3.73 × 10 −4 rad = 0.373 mrad
π w02 π (5.4 × 10−4)
(c) The spot size at the entrance to the beam expander is,
1/2 1/2
wEnt (z = 30) = w02 1 + (z/z02 )2 = (0.54 mm) 1 + 302/1.4642 = 11.08 mm
Assume that 30 m is in the far field (it is not according to our adopted criterion but the ratio
z/z02 = (30/1.464) is large enough that the far field formulas are reasonably good. For example using the
far field divergence angle to estimate wEnt gives,
wEnt ≈ z θFF = (30 m) 3.73 × 10−4 = 0.01119 m = 11.19 mm

which is a good approximation. The expanded beam has a radius of 10 × wEnt ≈ 111 mm = 11.1 cm.
(d) The Rayleigh range for the beam exiting the beam expander is
2
πω0BE π 0.111)2
z0BE = = m = 6.12 × 104 m
λ 632.8 × 10−9
Using Eq. (22 47),
2
f 2 (Z1 − f ) (0.1) (0.2 − 0.1)
Z2 = f + = 0.1 m + 2 m ≈ 0.1 m
(Z1 − f )2 + z0BE
2
(0.2 − 0.1)2 + (6.12 × 104
Using Eq. (22 46),
2 2 !2
0.2 π 11.1 × 10 −2

1 1 Z1 1 π w01 2 1 1
2 = 2 1− + 2 = 2 1 − 0.1 +
632.8 × 10−9
w02 w01 f f λ (11.1 × 10 m)
−2 (0.1 m)2
!
1 1 π 11.1 × 10 −2
2 ≈ ⇒ w02 = 1.81 × 10−7m = 0.18 µm
w02 (0.1 m)2 632.8 × 10−9
This spot size is less than λ and so one must question the validity of using the Gaussian beam analysis.
2
Since z0B E ≫ Z1 − f ) , the approximate formulas would work well.
2
27-12. (a), (b) Equations (22 47) and (22 46) with Z1 = 0 gives directly,
f 2( − f
2
f 2/z01

f
Z2 = f + 2 2
=f 1− 2 2
= 2/z 2
( − f ) + z01 1 + f /z 01 1 + f 01
" 2 #
2 2

1 1 1 π w01 2 1 1 π w01 1 z01
2
= 2
+ = 2
1 + = 2
1 +
w02 w01 f2 λ w01 f2 λ w01 f2
(c) For example, for a lens of focal length 20 cm and for light of wavelength 500 nm, Z2 ≈ f if,
f2 2 2
2
2 ≪ 1 ⇒ f ≪ π w01 /λ
z01
f 2 λ2 4
≪ w01
π2
4
1.01 × 10 −15 m4 ≪ w01
w01 > 1.783 × 10 −4 m
So for these parameters as long as w01 is suitable larger than about 0.2 mm, Z2 ≈ f .
158
27-13. The total power is the irradiance integrated over the transverse plane,
Z 2 Z 2π Z ∞
w0 2 2
Φtot = I dA = I0 dϕ e−2 ρ /w (z) ρ d ρ
A w(z) 0 0
Let u = 2 ρ2/w 2(z) ⇒ d u = 4 ρ dρ/w 2(z) ⇒ ρ dρ = du w 2(z)/4

2 2 Z ∞
w0 w (z) π
Φtot = 2 π I0 e−udu = w02 I0
w(z) 4 0 2
27-14. Using the change of variable introduced in the solutions to problem 27 13, the fraction of transmitted
power can be formed as
R 2a2/w 2 −u
Φ(a) e du 2
/w 2
F= = 0R
∞ −u = 1 − e−2a
Φtot 0
e du
27-15. Use the result of problem 27 14 and compare the powers transmitted with the iris all the way open and
opened to some radius a, then the ratio F of these powers is,
2
/w 2
F = 1 − e−2a
so that,

−2a2/w 2 1
e = 1 − F ⇒ − 2 a /w = ln(1 − F ) = − ln
2 2
1−F
√
2a
w= r
1
ln 1 − F
√ √
27-16. The collimated beam length is taken to be 2 z0 = 2 (π w02/λ). Also d = 4.5 ( 2 w0) so w0 = d/(4.5 2 ). Thus,
π d2

π
d2 = 1.458 × 10−5 m−1 d2

2 z0 = =
20.25 λ 20.25 (1.064 × 10 m
−6
(2 z0)d=1cm = 14.58 m
(2 z0)d=2cm = 58.3 m
(2 z0)d=3cm = = 131.2 m
(2 z0)d=5cm = 364.5 m
27-17. Equation (27 58) gives,

2 d
m
Hm(ξ) = ( − 1)meξ m e−ξ
2
dξ
0
0 ξ2 d

− ξ2 2 2
= ( − 1)0 eξ e− ξ = 1

H0(ξ = ( − 1) e e
dξ 0
√
1 ξ2 d

− ξ2

ξ2

− ξ2
2 2x
H1(ξ) = ( − 1) e e =−e −2ξe =2ξ =
dξ w
2
2 ξ2 d 2 d
2
2
2

H2(ξ) = ( − 1) e −ξ = ξ − 2 −ξ = ξ − 2 − ξ − 2 ξ − 2 ξ)e −ξ 2
2
e e ξ e e e (
dξ 2 dξ
√ !
2 x 8 x2
H2(ξ) = 4 ξ 2 − 2 = 4 2−2= 2 −2
w w
159
27-18. Given,
" 2 #
λ2 z 2 + π w02

i λz
p(z) = ln 2 − (m + n + 1) tan −1
2 π w02 π w02
Now, z0 = π w02/λ, so,

" 2 #
z 2 + π w02/λ

1 λz
i p(z) = − ln 2 − i(m + n + 1) tan −1
2 π w02/λ π w02

z + z02
2
1
i p(z) = − ln − i(m + n + 1) tan−1(z/z0)
2 z02
Recall that,
z 2 + z02 z 2 + z02 w 2(z)

w 2(z) = w02 1 + z 2/z02 = w02

⇒ =
z02 z02 w02
Then,
2
1 w (z)
i p(z) = − ln − i(m + n + 1) tan−1(z/z0)
2 w02
" −1/2 #
w 2(z)
i p(z) + i(m + n + 1) tan −1
(z/z0) = ln
w02

w0
i p(z) + i(m + n + 1) tan −1
(z/z0) = ln
w(z)
−1
(z/z0) w0
eip(z)+i(m+n+1)tan =
w(z)
w0 −i(m+n+1)tan −1(z/z0)
eip(z) = e
w(z)
27-19. The row appears as
2
2
2
H2(xs) H2 e−xs/2 H2 e−xs/2 Burn Pattern
H20(xs) 2
2 H202 e-xs
H20 e-xs /2
xs
xs
xs
160
Φ(a) 2 2
27-20. (a) For the TEM00 mode the fraction transmitted is given in problem 27 14 as F00 = = 1 − e−2a /w
Φtot
(b) For TEM01 : The irradiance is
2 2 √ 2 2
I01 = I0 w02/w 2 H02(xs) H12 (ys) e− ρ /w )=I0 w02/w 2 2 2 y/w e− ρ /w

Do the integrals in cylindrical coordinates with y = ρ sin ϕ. The fraction transmitted is

R 2π R a
Φ(a) I01 ρ dρ dϕ
F01 = = R 02π R 0∞
Φtot I01 ρdρ dϕ 0 0
Further make the change of variables used in problem 27 14: u = 2 ρ2/w 2(z) ⇒ ρ dρ = du w 2(z)/4
After canceling common terms, in the expression for F01 ,
R 2π R 2a2/w2
0
sin2 ϕ dϕ 0 u e−udu
F01 = R 2π
sin2 ϕ dϕ 0∞ u e−udu
R
0
I’ll use the free computer algebra system,
(%i1) T integrate((sin(j))^2,j,0,2*%pi)*integrate(u*exp(-u),u,0,2*a^2/w^2)/%pi/integrate(u*exp(-u),u,0,inf);
Is a zero or nonzero?nonzero
− 2a2
w2 + 2 a2 e w2
(%o7) F0 1 = 1 −
w2
2 2
Simplification gives, F01 = 1 − 1 + 2 a2/w2 e−2a /w

(c) Similarly, for the TEM11 mode with y = ρ sinϕ, x = ρ cosϕ, u = 2 ρ/w2
R 2π R ∞ √ 2 √ 2 2 2
e−2 ρ /w ρ dρdϕ R 2π sin2 ϕ cos2 ϕ dϕ R 2a /w u2 e−udu
2 2
0 0
2 2 y/w 2 2 x/w
0 0
F11 = R √ 2 √ 2 = R 2π
sin2 ϕ cos2 ϕ dϕ 0∞ u2 e−udu
2 2
R
2π R ∞
0 0
2 2 y/w 2 2 x/w e−2 ρ /w ρ dρ dϕ 0
2 !
2 a2 1 2 a2

2 2
Integration under Maxima and subsequent simplification gives, F11 = 1 − 1 + 2 + e−2a /w
w 2 w2
(d) The same procedure gives
R 2π R ∞ √ 2 2 2
4 ( 2 y/w)2 − 2 e−2 ρ /w ρ dρdϕ R 2π R 2a /w 16u2 sin4 ϕ − 8 u sin2 ϕ + 4 e−ududϕ
2 2
0 0 0 0
F02 = R √ 2 = R 2π R ∞
2π R ∞
4 ( 2 y/w) 2−2 e −2 ρ2/w 2
ρ dρdϕ 0 0
16u2 sin4 ϕ − 8 u sin2 ϕ + 4)e−ududϕ
0 0

1 2 2 2 2
Simplification gives, F02 = 1 − 2 2 2 a /w + 1 e−2a /w
2 2

2 a /w +
2 3
27-21.
1
0.8
Transm tted Fract on
0.6
0.4
TEM00
0.2 TEM01
TEM11
TEM02
0
0 0.5 1 1.5 2 2.5
a/w
161
27-22. From the transmittance curves in problem 27 21 it is evident that the transmittance of the TEM 00 mode is
significantly higher than that of the higher order modes for apertures a/w < 2. It is somewhat higher at all
finite aperture sizes. The spot size w is determined by the nature of the cavity and is the same for all of
the modes. One could place an adjustable aperture in the laser cavity and change the size of the aperture
until only the TEM00 mode has low enough loss to lase.
λ λ 488 × 10−9
27-23. (a) θFF = ⇒ w0 = = m = 1.55 × 10−4 m = 0.155 mm
π w0 π θFF π (0.001)
2 Φtot 10
(b) I0 = = W/m2 = 1.32 × 108 W/m2 = 1.32 × 104 W/cm2
π w02 π (1.55 × 10 −4)2
2
π w02 π 1.55 × 10−4
(c) The Rayleigh range is z0 = = m = 0.155 m. The irradiance is then
λ 488 × 10−9
z02 0.1552
I(ρ = 0, z = 10 m) = 1.32 × 104 W/cm2 = 3.17 W/cm2

I0 =
z + z02
2 102 + 0.1552
27-24. The Rayleigh range is needed:

The beam waist is, by symmetry, in the center of the cavity and so the z1 = − 0.1 m and z2 = 0.1 m. Also
R2 = 1 m. Then,
r
z02

R(z2)
R(z2) = z2 1 + 2 ⇒ z0 = z2 − 1 = 0.3 m
z2 z2
Equation (27 67) gives the cavity mode frequencies as

m+n+1
tan −1(z2/z0) − tan−1(z1/z0)

νmnq = q +
π
Using this relation,
3 × 108
(a) ν0,0,q+1 − ν0,0,q = c/2d = Hz = 750 MHz
0.4
3 × 108
(b) νm,n, q+1 − νm,n, q = c/2d = Hz = 750 MHz
0.4

c 1 −1 z2 −1 z1
(c) ν0,1, q − ν0,0,q = tan − tan = 154 MHz
2d π z z
0 0
c 1 z z
(d) ν1,0,q − ν0,0, q = tan−1 2 − tan−1 1 = 154 MHz
2d π z0 z
0
c 2 z z
(e) ν1,1, q − ν0,0,q = tan−1 2 − tan−1 1 = 308 MHz
2d π z0 z0
162
Chapter 28 Selected Modern Applications
Note: Many of the problems in this chapter are open-ended research problems. We do not provide
solutions to those problems.
28-1. (a) Excimer lasers emit ultraviolet electromagnetic radiation. The short wavelength emission can be
focused to a smaller spot size and so is useful in micro surgical applications where precise small cuts need
to be made. Recall Eq. 27 50 in which it is indicated that the spot size of a focused beam is directly pro
portional to the wavelength of the radiation.
(b), (c) Hemoglobin remains strongly absorbing in the UV range and the absorption coefficient of water
begins to increase below about 300 nm making excimer lasers useful as scalpels and for photo coagulation.
28-3. The Nd:YAG laser is typically used in this procedure. Possible design parameters are:
Wavelength: 1.06µm lens

Pulse width: 10 ns
Energy per pulse: 4 mJ w0 1 w0 2
Spot size of beam
incident on focusing 3.4 mm focal spot
lens
Power of focusing lens 20 D iris
f
Assume that the beam is incident on the lens at a waist of radius w01 = 3.4 mm. The Eq. (27 46) (with
Z1=0) give the focal spot radius as,
!
1 1 1 π w01 1 1 π (3.4 × 10−3 m
2 = 2 + f2 = +
1.06 × 10−6 m
⇒ w02 = 5 µm
w02 w01 λ (3.4 × 10−3)2 (0.05 m)2
The confocal parameter for the focused beam is,
2 π (5 × 10−6)2
z02 = π w02 /λ = m = 7.40 × 10−5 m = 0.074 mm
1.06 × 10−6
Twice this distance is the length of focal region of the focused spot.
28-4. During the pulse the power in the beam is

0.004 J
P =Φ= = 4.0 × 105 W
10−8 s
For a TEM00 beam, the irradiance at the center of the spot is given by problem 27 13 as,
2Φ 8 × 105
I= 2 = W/m2 = 1.0 × 1016 W/m2 = 1012 W/cm2
π w02 π (5 × 10−6)2
This value is in rough agreement with the estimate given in Example 28 1.
r
2 · 5.7 × 1015
r
ε0c 2 2I
28-5. I = E ⇒ E0 = = = 2.1 × 109 V/m
2 0 ε0 c 8.85 × 10−12 · 3 × 108
28-6. Using Eq. (27 48) one can estimate the beam diameter on the earth as,
D = f (2 θFF ) = (3.6205 × 107 m 5 × 10−6 = 180 m

The average irradiance over the beam is found roughly as, I = Φ/(π D2/4 = 7700 W/m2. Using the result

from problem 28 5,
r
2 · 7700
r
ε c 2I
E0 = I = 0 E02 ⇒ E0 = = V/m = 2400 V/m
2 ε0 c 8.85 × 10−12 · 3 × 108
163
28-7. (a), (b)
" 1/2 #
4 × 50

1 4P 1
RNHZ = −d = − 0.3 cm
d ϕ π MPE 0.003 π 5.1 × 10−4
Laser ϕ
RNHZ = 37, 100 cm = 371 m

RNHZ
28-8. (a), (b)
" 1/2 #
7.5 4 × 50

f 4P
RNHZ = = cm
d b π MPE 0.5 π 1.6 × 10−3
b ϕ ≈ b/f
Laser
RNHZ = 2990 cm = 29.9 m
f RNHZ
1 1 1 1 1
28-10. (a) ν p = νs + νi ⇒ νi = ν p − νs ⇒ = − = − ⇒ λi = 3610 nm
λi λ p λs 1060 nm 1500 nm
(b) The following two relations must both be satisfied,
d = ms λs/2 = miλi/2
d≈1m
with d being the cavity length. These conditions are met for, for example, by
d = 1.001775 m = 370 (3610) (1500 × 10−9 m /2 = 370 (1500) (3610 nm)/2

In practice one would need active control of the cavity length to maintain the dual resonance.
(c) If both beams are TEM00 modes the resonant frequencies are given by Eq. (27 67) as,

1 c
ν00 q = q + tan−1(z2/z0) − tan−1(z1/z0)
π 2d
tan−1(z2/z0)

c 1 λ 00 q
q + tan (z2/z0) − tan (z1/z0) =
−1 −1

d= q+
2 ν00 q π 2 π
Here I have noted that the beam waist is at the flat mirror at z1 = 0. The confocal parameter can be found
from the relation,
R2 = R(z2 = d) = d (1 + z02/d2)
−1/2
R2
−1 = d/z0 = z2/z0
d
Although d will not be exactly 1 m it is a good approximation to set it to the nominal value in the argu
ment inside the arctan function. Then,
−1/2
tan−1(z2/z0) tan−1(2/1 − 1) 1
≈ =
π π 4
Then the relations that must be satisfied for dual resonance are,

λs 1 λi 1
d= qs + = qi +
2 4 2 4
The cavity length could therefore be,
d ≈ 0.9991366875 m = 369 [ (3610) + 1/4](1500 × 10−9 m /2 = 369 [ (1500) + 1/4](3610 × 10−9 m /2

Again, one would not set the cavity length to this value but rather scan the cavity length until a dual reso
nance is reached and this is the length corresponding to one of the possible dual resonances.
164
28-11. If the field is resonant with the cavity the and there are no absorption losses all of the incident power is
transmitted through the cavity. In that case the intracavity power co propagating with the incident light
will be
Ptansm itted Pinc 0.1 W
Pcoprop = = = = 3.33 W
T (1 − R) 1 − 0.97
The counter propagating intracavity field will have a power of,
Pcounterprop = R Pcoprop = (0.97 ( 3.33 W) = 3.23 W

r
3 kB T
28-16. From thermodynamics: vrm s = .
m r
3 · 1.38 × 10−23 · 293
(a) For the oxygen molecule at T = 68 F = 20 C = 293 K, vrm s ≈
◦ ◦
m/s = 480 m/s
32 · 1.67 × 10−27
r r
3 kB T 3 · 1.38 × 10−23 · 0.002
(b) For rubidium at T = 0.002 K, vrm s = ≈ m/s = 0.8 m/s
m 85.5 · 1.67 × 10−27
28-17. As in the solution to problem 28 16b but with T = 0.003 K,

r r
3 kB T 3 · 1.38 × 10−23 · 0.003
vrms = ≈ m/s = 0.98 m/s
m 85.5 · 1.67 × 10−27
The detuning should match the Doppler shift. So,
∆ν ≈ ν vrm s /c = vrm s /λ = (0.98 m/s)/ 800 × 10−9 m = 1.225 × 106 Hz

165

Solution For Optics Pedrotti

Uploaded by

Copyright:

Available Formats

Solution For Optics Pedrotti

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Solution For Optics Pedrotti

Uploaded by

Copyright:

Available Formats

Chapter 1 Nature of Light

Energy n h ν n h c 100 6.63 × 10−34 J · s 3 × 108m/s

1-3. The energy of a photon is given by E = h ν = h c/λ

6.63 × 10−34 J · s 3 × 108m/s

6.626 × 10−34 J · s 2.998 × 108 m/s

b) λ = h/p = 6.63 × 10−34 J · s / 1.49 × 10−18 kg · m/s = 4.45 × 10−16 m

1000 W/m2 10−4 m2

c 3.00 × 108 m/s c 3.00 × 108 m/s

half wave antenna is then λ/2 = 1.5 m.

2-3. Refer to the figure for the relevant parameters.

h1 The person must be able to see the top of

mirror hm irror = h − h1/2 − h2/2 = h/2

h2 The mirror must be half the height of the

2-5. Refer to the figure.

2-8. Referring to the figure one can see that,

For the negative meniscus lens

For the back focal length: for lens 1:

2-23. The arrangement of the object and lenses is shown below.

2-27. See Figure 2 38 in the text. The applicable relations are:

2-32. Refer to Figure 2 39 in the text.

2-33. At each surface use the relation,

The image distance is then found as,

The image distance is then found as,

The image distance is then found as,

The image distance is then found as,

Image Position and Size:

The system is drawn to scale below.

(b) Image Plane:

(c) Candidates for AS: Lens1, Aperture, or Lens 2.

(d) Exit Pupil Ex P: Image of AS (i.e., Lens 1) in Lens 2:

(f) Angular field of view: Drawn not to scale,

3-5. Refer to Figure 3 5 in the text.

3-8. The necessary data is

4240 1.5286 − 1.5205

1.319 × 104 1.7328 − 1.7076

Then for a third wavelength λD = 589.2 nm,

(c) The resolving powers are,

RC = b DC = (7.5 × 107 nm) (4.146 × 10−5nm −1) = 3110

The minimum resolvable wavelengths are,

crown: (∆λ)m in = λ/RC = 5892 Å/3110 = 1.9Å

3-10. The indices of refraction of the lines are,

sin [(60 + 38◦20 )/2]

nC − = 1.51323. Similarly, nD = 1.51570 and nF = 1.52308

or A2 = 5.99◦. For the C and F lines with this double prism,

δF 2 = (1.536 − 1) (5.99) = 3.211◦ δF 1 = (1.648 − 1) (5) = 3.240◦ Net: δF 2 − δF 1 = 0.029◦

Dispersion = δF 12 − δC12 = 0.029 − ( − 0.007) = 0.036◦ = 2.16 ′

δC1 = (nC1 − 1) A1 = (1.5205 − 1) 15 = 7.8075◦

δF 2 − δC2 = 0.1215◦ = (nF 2 − nC2) A2 = (1.7328 − 1.7076) A2 ⇒ A2 = 4.82◦

3-14. Plots as a function of the different parameters are shown below.

A = 8, s0 = 2 m, d = 0.05mm f = 5 cm, s0 = 2 m, d = 0.05 mm A = 2, f = 5 cm, d = 0.05 mm

3-15. From Eqs. (3 30) and (3 31),

The ratio of the magnifications is

3-23. (a) The system is sketched below.

3-32. Using Eq. (3 35) and the given information,

(d) θ = DFL /L = 2/30 = (1/15) rad = 3.8◦.

3-34. Refer to figure 3 41 in the text. For M2:

The overall magnification is then, Moverall = m M = (5)( − 36) = − 180 ×

4-2. (a) Replace x by x + v t:

4-6. (a) y = A sin 2 π(x/λ ± t/T ) = (5 m)sin[2 π x/(50 m)]