Manual Introduction To Optics Pedrotti PDF

C hapter 1 Nature of Light
h h 6. 63 × 1 0 − 3 4 J · s
1 -1 . a) λ = = = = 6. 63 × 1 0 − 3 4 m
p m v ( 0. 05 kg) ( 20 m/ s)
h h 6. 63 × 1 0 − 3 4 J · s
b) λ = =√ = = 3. 88 × 1 0 − 1 0 m
p 2 m E [ ( 2 · 9. 1 1 × 1 0 − 3 1 kg) ( 1 0 · 1 . 602 × 1 0 − 1 9 J) ] 1 / 2

Energy n h ν n h c 1 00 6. 63 × 1 0 − 3 4 J · s 3 × 1 0 8 m/s
1 -2 . P = = = = = 3. 62 × 1 0 − 1 7 W
time t tλ ( 1 s) ( 550 × 1 0 − 9 m)
1 -3. The energy of a photon is given by E = h ν = h c/ λ

6. 63 × 1 0 − 3 4 J · s 3 × 1 0 8 m/s 1 eV
At λ = 380 nm: E = = 5. 23 × 1 0 − 1 9 J = 3. 27 eV
380 × 1 0− 9 m 1 . 60 × 1 0− 1 9J
− 34
6. 63 × 1 0 J · s 3 × 1 0 m/s
8 1 eV
At λ = 770 nm: E = = 2. 58 × 1 0 − 1 9 J = 1 . 61 eV
770 × 1 0 m − 9 1 . 60 × 1 0 − 1 9 J
h hc hc h
1 -4. p = E/ c = m c 2 / c = m c = 2. 73 × 1 0 − 2 2 kg · m/s, λ= = = = = 2. 43 × 1 0 − 1 2 m
p E m c2 m c
2 1 MeV
1 -5. Ev = 0 = m c 2 = 9. 1 09 × 1 0 − 3 1 kg 2. 998 × 1 0 8 m/ s = 8. 1 87 × 1 0 − 1 4 J = . 51 1 MeV
1 . 602 × 1 0 − 1 9 J
√ √
1 -6. c p = E 2 − m 2 c 4 , where E = EK + m c 2 = ( 1 + 0. 51 1 ) ) MeV. So c p = 1 . 51 1 2 − 0. 51 1 2 MeV
That is, c p = 1 . 422 MeV and p = 1 . 422 MeV/ c.

hc 6. 626 × 1 0 − 3 4 J · s 2. 998 × 1 0 8 m/ s 1 eV 1 Å 1 2, 400
1 -7. λ = = = Å · eV
E E 1 . 602 × 1 0 − 1 9 J 1 0− 1 0 m E
!
1 h − 1 /2 i 1
1 -8. EK = m c 2 p −1 = m c2 1 − v 2 / c2 − 1 ' m c 2 1 − ( − 1 / 2) v 2 / c 2 − 1 = m v 2
1 − v 2 / c2 2
1
1 -9. The total energy of the proton is,
−19

9 1 . 60 × 1 0 J 2
E = EK + m p c = 2 × 1 0
2
+ 1 . 67 × 1 0 − 2 7 kg 3. 00 × 1 0 8 m/s = 4. 71 × 1 0 − 1 0 J
1 eV
q h 2 2 2 i − 1 /2
E 2 − m 2p c 4 4. 71 × 1 0 − 1 0 J − 1 . 67 × 1 0 − 2 7 kg 3. 00 × 1 0 8 m
a) p = =
c 3. 00 × 1 0 8 m/ s
p = 1 . 49 × 1 0 − 1 8 kg · m/ s

b) λ = h/ p = 6. 63 × 1 0 − 3 4 J · s / 1 . 49 × 1 0 − 1 8 kg · m/ s = 4. 45 × 1 0 − 1 6 m

c) λ p h ot on = h c/ E = 6. 63 × 1 0 − 3 4 J · s 3. 00 × 1 0 8 m/s / 4. 71 × 1 0 − 1 0 = 4. 22 × 1 0 − 1 6 m

Energy Energy 1 000 W/m 2 1 0 − 4 m 2
1 -1 0. n ph ot on s = = = = 2. 77 × 1 0 1 7
hν h c/ λ ( 6. 63 × 1 0 − 3 4 J) ( 3. 00 × 1 0 8 m/ s) / ( 550 × 1 0 − 9 m)
n 1 E e / h ν 1 Ee λ 1 / h c λ 1
1 -1 1 . = = =
n 2 E e / h ν 2 Ee λ 2 / h c λ 2
1 -1 2 . The wavelength range is 380 nm to 770 nm. The corresponding frequencies are
c 3. 00 × 1 0 8 m/ s c 3. 00 × 1 0 8 m/s
ν7 7 0 = = = 3. 89 × 1 0 1 4 Hz ν3 8 0 = = = 7. 89 × 1 0 1 4 Hz
λ 770 × 1 0 m
− 9 λ 380 × 1 0 − 9 m

1 -1 3. The wavelength of the radio waves is λ = c/ ν = 3. 00 × 1 0 8 m/s / 1 00 × 1 0 6 Hz = 3 m. The length of the
half wave antenna is then λ/ 2 = 1 . 5 m.

1 -1 4. The wavelength is λ = c/ ν = 3. 0 × 1 0 8 m/ s / 90 × 1 0 6 Hz = 3. 33 m. The length of each of the rods is then
λ/ 4 = 0. 83 m.

1 -1 5. a) t = D l / c = ( 90 × 1 0 3 / 3. 0 × 1 0 8 s = 3. 0 × 1 0 − 4 s. b) D s = v s t = ( 340 ) 3. 0 × 1 0 − 4 m = 0. 1 0 m
Φe 500 W Φ 500 W
1 -1 6. a) Ie = = = 39. 8 W/ sr b) Me = e = = 1 0 6 W/ m 2
∆ω 4 π sr A 5 × 1 0 − 4 m2
Φ Φe 500 W
c) Ee = e = 2
= 2 = 9. 95 W/m 2 e) Φ e = Ee A = 9. 95 W/ m 2 π ( 0. 025 m) 2 = . 01 95 W
A 4πr 4 π( 2 m)
1 -1 7. a) The half angle divergence θ 1 / 2 can be found from the relation
r sp ot 0. 0025 m
tan( θ 1 / 2 ) ≈ θ 1 / 2 = = = 1 . 67 × 1 0 − 4 rad = . 0096 ◦
L ro om 15 m
2
A sp ot π r2 π ( 0. 0025 m)
b) The solid angle is ∆ ω = 2 = 2 s p ot = 2 = 8. 73 × 1 0 − 8 sr.
L ro om L ro om ( 1 5 m)
Φe Φe 0. 001 5 W
c) The irradiance on the wall is Ee = = 2 = = 76. 4 W/m 2 .
A s p ot π r sp ot π ( 0. 0025 m) 2
d) The radiance is ( approximating differentials as increments)
Φe 0. 001 5 W W
Le ≈ = = 8. 75 × 1 0 1 0 2
∆ω ∆A laser cosθ ( 8. 73 × 1 0 sr) π ( 0. 00025 m) cos( 0)
− 8 2 m · sr
2
C hapter 2 G eometrical O ptics
P P
d op ni xi
2 -1 . t = = i
c c
2 -2 . Referring to Figure 2 1 2 and with lengths in cm,
1 /2 1/2
n0 x2 + y2 + n i y 2 + s o + s i − x) 2 = no s o + ni s i
1 /2
2 1 /2
( 1 ) x2 + y2 + 1 . 5 y 2 + ( 30 − x) = 20 + 1 . 5 ( 1 0) = 35
1 /2 2
2. 25 y 2 + ( 30 − x) 2 = 35 − x 2 + y 2
1 /2
1 . 25 x 2 + y 2 + 70 x 2 + y 2 − 1 35 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 2 0. 4 20. 8 21 . 6 22. 4 2 3. 2 2 4. 0 24. 8 2 5. 6 2 6. 4 2 7. 2
y ( cm) 0 ± 1 . 0 ± 1 . 40 ± 1 . 96 ± 2. 69 ± 3. 2 0 ± 3. 58 ± 3. 85 ± 4. 04 ± 4. 1 4 ± 4. 1 8 ± 4. 1 3
2 -3. Refer to the figure for the relevant parameters.
√
d = d 0 = 30 2 + 2. 5 2 = 30. 1 04 cm d
t
d0
Fermat: d + d 0 = s + s 0 − t + m t
d + d0 = s + s 0 + t ( m − 1 )
2 ( 30. 1 0399) = 60 = t ( 1 . 52 − 1 )
t = 4 mm
s 0 = 3 0 cm s 0 = 30 cm
n = 1 . 52
2 -4. See the figure below. Let the height of the person be h = h 1 + h 2 .
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 h m irror = h − h 1 / 2 − h 2 / 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 0 ⇒ θ 0 = 30
30 ◦ √ √
At Side: 2 sin 60 ◦ = ( 1 ) sin θ 0 , sin θ 0 = 1 . 5 > 1
Side Thus total internal reflection occurs.
60◦
At Bottom: reverse of Top: θ 0 = 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 0 = or n = 0 = = 1 . 60
n s 1 . 87
7. 60/ 4
2 -7. Refer to Figure 2 33 in the text. By geometry, tan θ c = so θ c = 40. 1 8 ◦
1 2. 25
Snell’ s law: n sin θ c = ( 1 ) sin 90 ◦ ⇒ n = = 1 . 55
sin 40. 1 8 ◦
2 -8. Referring to the figure one can see that,

t
s = AB sin ( θ 1 − θ 2 ) and AB = . Therefore,
cos θ 2
t sin ( θ 1 − θ 2 ) θ1 A n1
s= . For t = 3 cm, n 2 = 1 . 50, θ 1 = 50 ◦ ,
cos θ 2 n 1
n2
Snell’ s law gives, sin θ 2 = 1 sin θ 1 = sin 50 ◦ . t θ1 − θ2
n2 1.5 θ2
3 sin ( 50 ◦ − 30. 71 ◦ )
Then, θ 2 = 30. 71 ◦ and s = = 1 . 1 53 cm.
cos 30. 71 ◦ B s
1 1 1
2 -9. Image of near end: s = 60 cm, = = , s 0 − 24 cm
60 s 0 − 40
1 1 1
Image of far end: s = 60 + 1 00 cm, + = , s 0 = − 32 cm. So, L 0 = ∆ s 0 = − 24 − ( − 32) = 8 cm
1 60 s 0 − 40
2-1 0. ( a) See Figure 2 34 in the text. Image due to rays directly from bubble through plane interface:
n1 n2 1.5 1
+ 0 = 0 or + 0 = 0 ⇒ s 0 = − 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 0 = − 7. 5 cm
2 s 10 R 2. 5 s 10 − 7. 5 1
n1 n2 1.5 1 0
refraction: + 0 = 0 or + 0 = 0 s 2 = − 1 0 cm
s s2 15 s2
Thus the images are at 3. 33 cm and 1 0 cm behind the interface.
2-1 1 . There are 5 unknowns: s 1 and s 10 in position ( 1 ) , s 2 and s 20 in position ( 2) , and the focal length f of the
mirror. The five equations that, solved simultaneously, yield the results are:
( 1 ) linear magnification: s 10 / s 1 = 2
( 2) linear magnification: s 20 / s 2 = 3
s s0 s s0
( 3) focal length from mirror equation: f = 1 1 0 ( 4) focal length from mirror equation: f = 2 2 0
s1 + s1 s2 + s2
( 5 ) image distance relation: s 20 = s 10 + 75
One finds s 1 = 1 1 2. 5 cm, s 2 = 1 00 cm, s 10 = 225 cm, s 20 = 300 cm, f = 75 cm
2-1 2 . The object distance from the front surface is the diameter of the sphere, 5 cm. Then,
0
n1 n2 n2 − n1 1.5 1 1 − 1.5 0 n s ( 1 . 5) ( − 1 0)
+ 0 = ⇒ + 0= ⇒ s = − 1 0 cm and, m = − 1 = − = + 3.
s s R 5 s − 2. 5 n2 s ( 1 ) ( 5)
n1 n2 n2 − n1
2-1 3. Generally, + 0 =
s s R
n1 n2 n2 − n1 n1 R
( a) + = or f = ( b) n 2 > n 1 : then R > 0 ( convex) , n 2 < n 1 : then R < 0 ( concave)
f ∞ R n2 − n1
4
n1 n2 n2 − n1 n s0
2-1 4. ( a) In this position the object distance is s = 1 5 cm. Then, using, + 0 = and m = − 1 ,
s s R n2 s
4 1 1 − 4/ 3 ( 4/ 3) ( − 1 5)
+ = ⇒ s = − 1 5 cm ( center) and, m = −
0
⇒ m = 4/ 3.
3 ( 1 5) s 0 − 15 ( 1 ) ( 1 5)
( b) Similarly, in this position s = 7. 5 cm so that,
4 1 1 − 4/ 3 n s0 ( 4/ 3) ( − 45/ 7)
+ 0= ⇒ s 0 = − 6. 4 cm, m = − 1 = − = 8 / 7.
3 ( 1 5/ 2) s − 15 n2 s ( 1 ) ( 1 5/ 2)
2-1 5. 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 s0 ( 1 ) ( 40)
( a) 1 + 20 = 2 ⇒ + 0= ⇒ s 0 = 40 cm. Then, m = 1 = =−1
s s R 30 3s 5 n2 s ( 4/ 3) ( 30)
( b) s = 25 − 40 = − 1 5 cm ( virtual object) , s 0 = − s = 1 5 cm, m = − s 0 / s = 1
4/ 3 1 1 − 4/ 3 − ( 4/ 3) ( − 1 5)
( c) + 0= ⇒ s 0 = − 1 5 cm. Then, m = = + 2.
10 s −5 ( 1 ) ( 1 0)
The overall magnification is m = ( − 1 ) ( + 1 ) ( + 2) = − 2. Thus a virtual, inverted, double sized image
appears 1 5 cm behind ( right) the spherical window.
2-1 6. The plane side of the lens has R 1 = ∞ . The radius of curvature R 2 of the convex side is then found from
the lensmaker’ s equation:

1 n2 − n1 1 1 1 1 . 52 − 1 1 1
= − ⇒ = − ⇒ R 2 = − 1 3 cm
f n1 R1 R2 25 1 ∞ R2

1 n − n1 1 1
2-1 7. In general the lensmaker’ s equation gives, = 2 −
f n1 R1 R2
For the positive meniscus
lens
shown to the right, R 1 = 5 cm and R 2 = 1 0 cm.
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, R 1 = 1 0 cm and R 2 = 5 cm.
1 1 . 50 − 1 1 1 R1
For this case, = − ⇒ f = − 20 cm
f 1 10 5 R2
2-1 8. 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
2 0 cm + 0 = , left lens surface
s1 s1 R1
n2 n2 n3 n3 − n2
F ish Ta n k
+ 0 = , right lens surface
n1
n3 s2 s2 R2
n1 n3 n2 − n1 n3 − n2
For a thin lens, s 2 ≈ − s 10 . Adding the equations, + 0 = + . Or, simply,
s1 s2 R1 R2
n1 n3 n2 − n1 n3 − n2 4 1 3/ 2 − 4/ 3 1 − 3/ 2
+ 0 = + ⇒ + = + , so that s 0 = − 22. 5 cm.
s s R1 R2 3 ( 20) s 0 30 − 30

n s0 n 2 s 20
The total magnification is m T = m 1 m 2 = − 1 1 − , where s 2 = − s 10 . So,
n2 s n3 s 2
n s0 ( 4/ 3) ( − 22. 5)
mT = − 1 = − = 1 . 50.
n3 s ( 1 ) ( 20)
5
1 1 1 1 1 1
2-1 9. ( a) Using = + as in Eq. ( 2 33) , = + or feq = 6. 67 cm
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 s 2 = f2 , s 2 = L − s 10 or s 10 = L − f2 .
s 2 ∞ f2
1 1 1 f ( L − f2 ) ( − 5 ) ( 1 0 − 20)
Lens 1 : + = or feq, f = 1 = cm = − 1 0 cm
feq, f L − f2 f1 L − ( f1 + f2 ) 1 0 − ( − 5 + 20)
For the back focal length: for lens 1 :
1 1 1 1 1 1
+ = ⇒ + = ⇒ s 10 = f1 . Then s 2 = L − f1 , so that, for lens 2:
s 1 s 10 f1 ∞ s 10 f1
1 1 1 1 1 1 f2 ( L − f1 ) ( 20) ( 1 0 − ( − 5 ) )
+ = ⇒ + = ⇒ feq, b = = cm = − 60 cm
s 2 s 20 f2 L − f1 feq, b f2 ( L − f1 ) − f2 ( 1 0 − ( − 5 ) ) − 20
2-2 0. 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 = 30 cm,
feq f1 f2 f3 f1 ∞ − 15

1 1 1 1 50 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 = − 50 cm.
feq 30 1 50 30
2-2 1 . ( a) One can use the formula derived in problem 2 1 9b, or do the calculation at first hand:
1 1 1 1 1 1
Second lens: + = or s 2 = 20 cm, First lens: + = or s 1 = 3. 33 cm. 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
2-2 2 . Refer to Figure 2 37 in the text.
1 1 1 s0 − 3 f/ 5
( b) Lens heading towards mirror: + 0= or s 0 = − 3 f/ 5 . m 1 = − = − = 2/ 5
3 f/ 2 s −f s 3 f/ 2
Mirror:
5 1 1 s0
s = 3 f + 3 f/ 5 = 1 8 f/ 5 ⇒ + 0 = ⇒ s 0 = 1 8 f/ 1 3, m 2 = − = − ( 1 8 f/ 1 3) / ( 1 8 f/ 5 ) = − 5/ 1 3
18 f s f s
Lens after reflection:
13 1 1 13
s = 3 f − 1 8 f/ 1 3 = 21 f/ 1 3 ⇒ + = or s 0 = 21 f/ 34, m 3 = − s 0 / s = − ( 21 f/ 34) / ( 21 f/ 1 3) =
21 f s 0 − f 34

2 5 13
mT = − = − 1 7. The image is inverted, ( 21 / 34) f behind ( right of) lens, inverted, and
5 13 34
1 / 1 7 original size.
2-2 3. The arrangement of the object and lenses is shown below.
20 cm 30 cm 20 cm
1 2 3
( a) f1 = + 1 0 cm, f2 = + 1 5 cm, f3 = + 20 cm
1 1 1
1 st lens: + = s 0 = 20 m 1 = − 20/ 20 = − 1
20 s 0 1 0
1 1 1
2nd lens: + = s 0 = − 30 m 2 = − ( − 30) / 1 0 = + 3
10 s0 15
1 1 1
3rd lens: + = s 0 = 1 00/ 3 m 3 = − 1 00/ 3 ( 50) = − 2/ 3
50 s 0 20
mT = m1 m2 m3 = + 2
( b) f1 = + 1 0 cm, f2 = − 1 5 cm, f3 = + 20 cm
1 1 1
1 st lens: + = s 0 = 20 m 1 = − 20/ 20 = − 1
20 s 0 1 0
1 1 1
2nd lens: + = s0= − 6 m 2 = − ( − 6 ) / 1 0 = + 0. 6
10 s0 − 15
1 1 1 10
3rd lens: + = s 0 = 520/ 6 m 3 = − 520/ ( 6 × 26) = −
26 s 0 20 3
mT = m1 m2 m3 = + 2
( c) f1 = − 1 0 cm, f2 = + 1 5 cm, f3 = − 20 cm
1 1 1 1
1 st lens: + = s 0 = 20/ 3 m 1 = − ( − 20) / 3 ( 20) =
20 s 0 − 1 0 3
3 1 1 ( 330) ( 3) 9
2nd lens: + = s = 330/ 1 3 m 2 = −
0
=−
110 s0 15 ( 1 3) ( 1 1 0) 13
− 13 1 1 ( 1 40) ( 1 3) 26
3rd lens: + 0= s = 1 40/ 1 9 m 3 = −
0
=
70 s − 20 ( 1 9) ( − 70) 1 9
m T = m 1 m 2 m 3 = − 6/ 1 9

1 n − n1 1 1
2-2 4. 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 − n L 1 1
and in the liquid: − = − .
1 88 nL R1 R2
− 1 88 0. 5n L
Dividing the two equations gives, = or n L = 1 . 63.
30 1 . 5 − nL
7
2-2 5. Use the lensmaker’ s formula to find the focal length of the lens,

1 n2 − n1 1 1 1.5 − 1 1
= − = 0+ ⇒ f = 1 20 cm
f n1 R1 R2 1 60
f x0
The Newtonian equations are, m = − = − . For m = − 4, .
x f
f 1 20
− 4=− = − or x = 30 cm
x x
x0 x0
− 4= − =− or x 0 = 480 cm
f 1 20
Thus, s = x + f = 30 + 1 20 = 1 50 cm and s 0 = x 0 + f = 480 + 1 20 = 600 cm.
1 1 1 1 1 1
Check: + = + = =
s s 0 1 50 600 1 20 f
1 1 1
2-2 6. ( a) f1 = 1 0 cm ⇒ P1 = = + 1 0 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 = 1 0 + 5 − 2. 5 = + 1 2. 5 D
1 1 1 0 1 1
( b) + 0 = , V + V = P , where V = = = + 8. 33 D,
s s f s 0. 1 2
0 1 1
V = 4. 1 67 D or s 0 = 0 = = 0. 24 m = 24 cm
V 4. 1 67
2-2 7. See Figure 2 38 in the text. The applicable relations are:

1 1 1 1 1 1
Lens equations: + 0 = and + 0 = ,
s1 s1 f s2 s2 f
Geometrical: L = s 1 + s 10 = s 2 + s 20 , D = s 2 − s 1 = s 10 − s 20
Thus,
s 1 s 10 s s0 s s0 s s0
f= 0 = 1 1 = 2 20 = 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,
s 2 = s 10 and s 20 = s 1 ( 2)
Adding and subtracting the equations L = s 2 + s 2 and D = − s 1 + s 2 , we get,
L − D = 2s and L + D = 2s 2 . Their product is by Eq. ( 1 ) , L 2 − D 2 = 4 s 1 s 2 , or by Eq. ( 2) , L 2 − D 2 = 4 f L.
L2 − D2
Thus, f = .
4L
1 1 1 1 1 1
2-2 8. Lens equations: + = and + = . Then calculate,
s 1 s 10 f s 2 s 20 f
1 1 s s s1 s2 s − f s1 − f s2 − s1
− = − 10 + 20 = − + = 2 − =
M1 M2 s1 s2 s 1 f/ ( s 1 − f) s 2 f/ ( s 2 − f) f f f
Thus,
s2 − s1
f=
1 / M1 − 1 / M2
2-2 9. 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 = a 2 + x 2 + b 2 + ( d − x) 2
A
B dD x − ( d − x)
θi θr =√ +p =0
a a d x a 2 + x2 b 2 + ( d − x) 2
x
sin θ i − sin θ r = 0 ⇒ θ i = θ r
d
8
2-30. The two set ups are illustrated below,
1 n n− 1 nR
Refraction at curved side: + = ; s0−
∞ s0 R n− 1
nR
Reflection at plane side: s 0 = − s = −
n− 1
n 1 1−n
Refraction at curved side: + = or s 0 = R/ 2 ( n − 1 )
− n R/ ( n − 1 ) s 0 −R
R
Thus, f1 =
2 (n − 1)
1 1 2 R
Reflection at curved face: + =− ; s0= −
∞ s0 R 2
n 1 R
Refraction at plane face: + = 0; s = 0
− R/ 2 s 0 sn
n 1 R
Refraction at plane face: + 0 = 0; s 0 =
− 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 0 = 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 = 2 f
ds ( s − f) 2
f ( 2 f)
and has the value D = 2 f + = 4f. That is, in this configuration s = s 0 = 2 f.
2f − f
2-32 . Refer to Figure 2 39 in the text.

( a) Let the angle with the normal to the interface in each region of index of refraction n i be θ i . Then
applying Snell’ s law sequentially at each interface leads to,
n 0 sinθ 0 = n 1 sinθ 1 = n 2 sinθ 2 . . . = n i sinθ i . . . = n f sinθ f
That is,
n 0 sinθ 0 = n f sinθ f
( b) In each medium the lateral displacement is t i tanθ i . The total lateral displacement y due to N media
can be written as,
X n
y= t i tanθ i
i=1
where sinθ i = ( n 0 / n i ) sinθ 0 .
2-33. At each surface use the relation,

n1 n2 n2 − n1
+ 0 =
s s R
For light incident first on the plane side:
1.5 1 1 − 1.5
1 st surface ( plane) : no change, 2nd surface ( curved) : + 0= or s 0 = 8 cm.
∞ s −4
For light incident first on the curved side:
1 1.5 1.5 − 1
1 st surface ( curved) : + 0 = or s 0 = 1 2 cm.
∞ s 4
1.5 1 1 − 1.5
2nd surface ( plane) : object distance = 4 − 1 2 = − 8 cm ( virtual) , + = or s 0 = 5. 33 cm
− 8 s0 ∞
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
+ 0 = ⇒ 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 = + 40 cm
f 10
1 1 − 4/ 3
For light incident from the medium of index 4/3: = or f = + 30 cm
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 = − 30 cm
∞ f 10
1 4/ 3 4/ 3 − 1
For light incident from the medium of index 1 : + = or f = − 40 cm
∞ f − 10
s 1 f 6 in
2-35. | m | = = = = , since s 0 = f. So s = 50, 000 × 6 in = 25, 000 ft
s 0 50, 000 s s
2-36. Using, the lensmaker’ s equation the focal power of the cylindrical lens is,
1 n 2 − n 1 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 0 = 1 8. 73 cm
s 0 f s 8. 33 cm 1 5 cm
Then, Eq. ( 2 37) gives,
s + s0 1 5 + 1 8. 73
AB = CL = 7 cm = 1 5. 75 cm
s 15
The line image is real, 1 8. 75 cm past the lens and 1 5. 75 cm long.
1 n 2 − n 1 1 . 52 − 1 . 0
= = ⇒ f = 28. 85 cm
f R 1 5 cm
1 1 1 1 1
The image distance is then found as, = − = − ⇒ s 0 = − 65. 2 cm
s 0 f s 28. 85 cm 20 cm
s + s0 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 − 1 0 cm
1 1 1 1 1
= − = − ⇒ s 0 = − 1 1 . 1 1 cm
s 0 f s − 20 cm 25cm
s + s0 25 − 1 1 . 1 1
AB = CL = 5 cm = 2. 78 cm
s 25
The line image is virtual, 1 1 . 1 1 cm from the lens on the object side of the lens and 2. 78 cm long.
1 n2 − n1 1 . 5 − 1 . 0
= = ⇒ f = − 40 cm
f R − 20 cm
1 1 1 1 1
= − = − ⇒ s 0 = − 1 3. 33 cm
s 0 f s − 40 cm 20cm
s + s0 20 − 1 3. 33
AB = CL = 2 cm = 0. 667 cm
s 20
The line image is virtual, 1 3. 33 cm from the lens on the object side of the lens and 0. 67 cm long.
1 n 2 − n 1 1 . 60 − 1 . 0
= = ⇒ f = 8. 33 cm
f R 5 cm
1 1 1 1 1
= − = − ⇒ s 0 = − 21 . 45 cm
s 0 f s 8. 33 cm 6 cm
s + s0 6 − 21 . 45
AB = CL = 7 cm = − 1 8. 0 cm
s 6
The line image is virtual, 21 . 45 cm from the lens on the object side of the lens and 1 8. 0 cm long.
11
C hapter 3 O ptical 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 0 = − 3. 33 cm
s s0 f 2 cm s 0 5 cm
s0 3. 33 5 5
m= − = − = ⇒ size = × 2 cm = 3. 33 cm
s 2 3 3
Image Position and Size:
1 1 1
+ = ⇒ s 0 = 1 0 cm
1 0 cm s 0 5 cm
m = − s 0 / s = − 1 0/ 1 0 = − 1 ⇒ size = 1 × 2 cm = 2 cm, inverted
The system is drawn to scale below.

Ex P AS L ens
Image
Ob ject
3-2 . Exit Pupil: AS ( no elements follow) , Entrance Pupil: Image of AS formed by lens
1 1 1
Position of En P : + = ⇒ s 0 = − 4. 29 cm = − 30/ 7 cm ( right of lens)
2. 5 cm s 0 6 cm
− 30 2 1 2 12
Size of En P: m = − s 0 / s = − × = size = × 2 cm = 3. 43 cm
7 5 7 7
1 1 1 3
Image: + = ⇒ s 0 = 1 0. 5 cm, m = − s 0 / s = − 1 0. 5/ 1 4 = − 0. 75; size = × 4 cm = 3 cm, inverted
1 4 cm s 0 6 cm 4
En P
AS
L ens
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 0 = − 1 2 cm. m = − s 0 / s = − = + 3, size = 3 × 2 cm = 6 cm
4 cm s 0 6 cm 4
1 1 1
Image: + = ⇒ s 0 = 1 0. 5 cm , m = − s 0 / s = − 1 0. 5/ 1 4 = − 0. 75, size = 3/ 4 × 2 = 1 . 5 cm
1 4 cm s 0 6 cm
Ex P
AS
L ens
Image
Ob ject
3-4. ( a) The vertical and horizontal scales in the figure differ by a factor of 4.
Ob ject plane Image plane
L ens 1 L ens 2
Aperture
( b) Image Plane:
f1 s 1 ( 40/ 3) ( 40) f s ( 20/ 3) ( 1 0)
s 10 = = = 20 cm ( right of Lens 1 ) s 20 = 2 2 = = 20 cm ( right of Lens 2)
s 1 − f1 40 − 40/ 3 s 2 − f2 1 0 − 20/ 3
( c) Candidates for AS: Lens1 , Aperture, or Lens 2.

2
Lens subtends at an angle of θ L 1 = = 0. 05 rad.
40
20 ( 40/ 3)
Aperture image in Lens 1 : s 0 = = 40 cm, or at object plane, then θ A p ert u re = 0 ◦
20 − 40/ 3
0 30 ( 40/ 3)
Lens 2 image through Lens 1 : s = = 24 cm left of Lens 1 or 1 6 cm right of object.
30 − 40/ 3
0 1.6
m = − s / s = − 24/ 30 = − 0. 8 so size = 0. 8 × 2 = 1 . 6 cm, θ L 2 = = 0. 1 > 0. 05
16
Thus Lens 1 behaves as the AS. It is also the En P, being the first in line.
( d) Exit Pupil E x P: Image of AS ( i. e. , Lens 1 ) in Lens 2 :

0
0 sf2 30 ( 20/ 3) s 8. 57 2 4
s = = = 8. 57 cm, right of Lens 2 , m = − = − = 0. 2857 = 2/ 7 so D E x p = × 2 = cm
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 En P= Lens
1:
0. 5 2
θ A p ert u re = = 0. 025; θ L en s 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.
0 ( 1 0) ( 20/ 3)
s = = 20 cm, right of Lens 2, in image plane,
1 0 − 20/ 3
0
m = − s / s = 20/ 1 0 = − 2 so D E x ω = 2 × 0. 5 = 1 . 0 cm
( f) Angular field of view: Drawn not to scale,
En W En P Ex P Ex W
1 1
1 cm α α= = 0. 01 25 rad α0 1 cm α0 = = 0. 0875 rad
40 1 0 − 8. 43
α = 1 . 43 ◦
α 0 = 5. 01 ◦
40 cm 1 1 . 43 cm
3-5. Refer to Figure 3 5 in the text.

( a) First determine whether A , L 1 , or L 2 ( its image though L 1 ) subtends the smallest angle at he center of
En P:
Rim of A at En P: 90 ◦ . Rim of L 1 at En P: tan − 1 ( 3/ 3) = 45 ◦ .
Rim of the image of L 2 in L 1 :
This image is labeled L 20 in Figure 3 5 in the text and is shown there to be virtual, erect, located 1 5 cm
from En P with half size 9 cm. Thus L 20 subtends an angle of tan − 1 ( 9/ 1 5) = 31 ◦ at En P. So L 20 subtends
the smallest angle at the entrance window making L 2 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 L 2 , L 20
is the entrance window. As shown in Figure 3 5 in the text it is located 1 2 cm to the right of L 1 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 ( L 2 ) , the exit window is L 2 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 L 1 the rays are head towards
P 0 . Before they reach P 0 they are redirected towards the tip of the final image at P 00 . The backward exten
sion of the marginal rays heading toward P 00 skim the edges of the exit pupil.
3-6. The deviation δ is given as a function of the input angle θ 1 by Eqs. ( 3 7) , ( 3 8) , ( 3 1 0) , and ( 3 1 1 ) as,
δ = θ 1 + θ 2 − θ 10 − θ 20 = θ1 + θ 2 − A
sin( θ 1 )
δ = θ 1 + sin − 1 n sin A − sin − 1 −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 1 5)

A + δm
sin 2 sin ( 30 − δm / 2)
n= =
sin ( A/ 2) 1 /2
1
For n red = 1 . 525 sin ( 30 + δm / 2) = ( 1 . 525) or δm , red = 39. 37 ◦
2
1
For n b lu e = 1 . 535 sin ( 30 + δm / 2) = ( 1 . 535) or δm , b lu e = 40. 26 ◦
2
δm , b lu e − δm , red = 0. 89 ◦ = 53 0
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 λ F and λ C are: n F = A + B/ λ 2F , n C = A + B/ λ 2C . Solving these simultaneously,
B nF − nC
A = nF − and B =
λ 2F ( 1 / λ 2F − 1 / λ 2C )
4240 1 . 5286 − 1 . 5205

For crown glass: A = 1 . 5286 − = 1 . 51 1 B= = 4240 nm 2
466. 1 2 1
−
1
( 48 6 . 1 ) 2 ( 656. 3) 2
1 . 31 9 × 1 0 4 1 . 7328 − 1 . 7076
For flint glass: A = 1 . 7328 − = 1 . 677; B = = 1 . 31 9 × 1 0 4 nm 2
486. 1 2 1
2 −
1
( 486. 1 ) ( 656. 3)
2
Then for a third wavelength λ D = 589. 2 nm,
B 4240
Crown: n D = A + 2 = 1 . 51 1 + ( 589. 2) 2 = 1 . 523 , ( table: 1 . 5230)
λD
1 . 31 9 × 1 0 4
Flint: n D = 1 . 677 = = 1 . 71 5 , ( table: 1 . 7205)
( 589. 2) 2
dn 2B
( b) The dispersion is D = = − 3 so,
dλ λ
− 2( 4240 nm 2 ) − 2 ( 1 31 90 nm 2 )
DC = = − 4. 1 46 × 1 0 − 5 nm − 1 D F = = 1 . 290 × 1 0 4 nm − 1
( 589. 2 nm) 3 ( 589. 2 nm) 3
( c) The resolving powers are,
R C = b D C = ( 7. 5 × 1 0 7 nm) ( 4. 1 46 × 1 0 − 5 nm − 1 ) = 31 1 0
R F = b D F = ( 7. 5 × 1 0 7 nm) ( 1 . 290 × 1 0 − 4 nm − 1 ) = 9675
The minimum resolvable wavelengths are,
crown: ( ∆λ) m in = λ/ R C = 5892 ˚A / 31 1 0 = 1 . 9A

˚
flint: ( ∆ λ) m in = λ/ R F = 5892 ˚A / 9675 = 0. 61 ˚A
15
sin ( 30 + δ/ 2)
3-9. ( a) 1 . 6381 = or δ = 50 ◦
sin 30
n F − n C 1 . 63461 − 1 . 6451 1 1
( b) ∆ = = =−
nD − 1 1 . 6381 − 1 55. 5
n1 − n2
( c) D = − 2B/ λ 3 where B = 1 1 , as in problem 6 8a. For the long wavelength region, take λ 1 = 587. 6 nm
λ 21
− λ2
2
1 . 6381 0 − 1 . 63461 B 6073. 7
and λ 2 = 656. 3 nm. Then B = = 6073. 7 nm 2 and A = n 1 − 2 = 1 . 6381 0 − = 1 . 6205.
1 / 587. 6 2 − 1 / 656. 3 2 λ 587. 6 2
− 2 ( 6073. 7)
Finally, D ( at λ = 656. 3 nm) = = 4. 297 × 1 0 − 5 nm − 1
656. 3 3
λ
( d) The resolving power is R = AV = b D . Now ∆λ = 656. 2852 − 656. 271 6 = 0. 01 36 nm, so that
∆λ
λ AV 656. 28 nm
b= = = 1 . 1 2 × 1 0 nm = 1 . 1 2 m!
D ∆ λ ( 4. 297 × 1 0 − 5 nm − 1 ) ( 0. 01 36 nm)
3-1 0. The indices of refraction of the lines are,
0
sin [ ( 60 + 38 ◦ 20 ) / 2]
nC − = 1 . 51 323. Similarly, n D = 1 . 51 570 and n F = 1 . 52308
sin ( 60/ 2)
nF − nC
Then, ∆ = = 0. 01 909
nD
3-1 1 . Let the flint prism be prism ( 1 ) and the crown be prism ( 2) . Then, dispersion of the flint prism is δD 1 =
◦ ◦
( 1 . 635 − 1 ) ( 5 ◦ ) = 3. 1 75 . This dispersion must be matched by the crown prism, δD 2 = 3. 1 75 = ( 1 . 53 − 1 ) A 2 ,
or A 2 = 5. 99 ◦ . For the C and F lines with this double prism,
δF 2 = ( 1 . 536 − 1 ) ( 5. 99) = 3. 21 1 ◦ δF 1 = ( 1 . 648 − 1 ) ( 5 ) = 3. 240 ◦ Net: δF 2 − δF 1 = 0. 029 ◦

δC 2 = ( 1 . 527 − 1 ) ( 5. 99) = 3. 1 57 ◦ δC 1 = ( 1 . 6301 − 1 ) ( 5) = 3. 1 50 ◦ Net: δC 2 − δC 1 = − 0. 007 ◦
Dispersion = δF 1 2 − δC 1 2 = 0. 029 − ( − 0. 007) = 0. 036 ◦ = 2. 1 6 0
3-1 2 . Let the crown prism be prism 1 and the flint prism be prism 2. Then,
δC 1 = ( n C 1 − 1 ) A 1 = ( 1 . 5205 − 1 ) 1 5 = 7. 8075 ◦
δF 1 = ( n F 1 − 1 ) A 1 = ( 1 . 5286 − 1 ) 1 5 = 7. 929 ◦
The relative dispersion is δF 1 − λ C 1 = 0. 1 21 5 ◦ . This must be reversed by the flint prism:
δF 2 − δC 2 = 0. 1 21 5 ◦ = ( n F 2 − n C 2 ) A 2 = ( 1 . 7328 − 1 . 7076) A 2 ⇒ A 2 = 4. 82 ◦
The deviation of line D is then, δD 1 − δ D 2 = ( n D 1 − 1 ) A 2 = ( . 523) ( 1 5) − ( 0. 7205) ( 4. 821 ) = 4. 37 ◦
16
Φ t ot 25 W
3-1 3. ( a) The radiant exitance is given by Me = = = 1 0 4 W/ m 2 .
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 Z2π Z π/ 2
1
Φ t ot = Ie dω = dϕ Ie ( 0) cos( θ) sin( θ) dθ = 2 π I( 0) = π Ie ( 0)
0 0 2
Thus,
Φ t ot 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 L e is independent of the viewing angle θ and given by,
I( 0) cos θ I( 0) 7. 96 W/ sr W
Le = = = = 31 80 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 Φ c am 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
Φ c am = I( 0) ∆ωlen s = I( 0) ( A len s ) / r 2 = ( 7. 96 W/ sr) = 1 . 56 × 1 0 − 4 W
( 1 00 cm) 2
Φ c am Φ
( c) The irradiance at the film is then, Ee = = c am
A im ag e | m | 2 A
Here, we have noted that each image dimension is decreased from the source dimension by the magnifica
tion,
s0 f 4
| m| = = = = 0. 041 7
s s − f 1 00 − 4
So,
1 . 56 × 1 0 − 4 W
Ee = = 35. 9 W/ m 2
0. 041 7) 2 ( 0. 05 m) 2
3-1 4. 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 , s 0 = 2 m, d = 0. 05 mm f = 5 cm, s 0 = 2 m, d = 0. 05 mm A = 2, f = 5 cm, d = 0. 05 mm
3-1 5. From Eqs. ( 3 30) and ( 3 31 ) ,

A s 0 d ( s 0 − f) ( 4) ( 6 ) ( 1 0 − 6 ) ( 6 − 0. 05)
s0 − s1 = = m = 0. 057 m = 5. 7 cm
f2 + A s0 d ( . 05) 2 + ( 4) ( 6 ) ( 1 0 − 6 )
A s 0 d ( s 0 − f)
s2 − s0 = ≈ s 0 − s 1 = 5. 7 cm
f2 − A s 0 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
3-1 6. Using Eq. ( 3 35) ,
1 1 1 L 1 1 15
= + − = + − ⇒ f = 53. 3 cm
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 s 2 = − 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
+ 0 = ⇒ + = ⇒ s 20 = 1 3. 33 cm
s 2 s 2 f2 − 5 s2 − 8
That is, the film should be placed 1 3. 33 cm past the negative lens of the combination. The image size h 0 of
a distant object subtending an angle of 2 ◦ at the camera is found using the effective focal length of the lens.
| h 0| h0
tan 2 ◦ = = ⇒ 1 . 86 cm
f 53. 33 cm
Alternately one may find the image height as,

s0 s0 20 cm 1 3. 33 h
h 0 = m1 m2 h = − 1 − 2 = h = ( 53. 33 cm) = ( 53. 33 cm) tan 2 ◦ = 1 . 86 cm
s1 s2 s1 −5 s1
3-1 7. Using the parameters f = 5 cm = 0. 1 640 ft, s 0 = 6 ft, A = f/ 4 = 4, and d = 0. 05 mm = 1 . 640 × 1 0 − 4 ft in Eqs,
( 3 30) and ( 3 31 ) gives,
s 0 f ( f + A d)
s1 = = 5. 26 ft
f2 − A d s0
s f ( f − A d)
s2 = 0 2 = 7. 00 ft
f − A d S0
3-1 8. See Figure 3 38 in the text. The diameter of the image of the sun is about

π rad
D I = fL θ = ( 50 cm) ( 0. 5 ◦ ) = 0. 44 cm. Since all of the power intercepted by the lens, falls on the
1 80 ◦
image, the irradiance of the image is
2
A im a ge D L2 5
Iim age = Is u n = Isu n 2 = 1 000 W/cm 2
= 1 . 29 × 1 0 5 W/ cm 2
A len s DI 0. 44
3-1 9. ( a) Using f = 1 5 cm, s = 1 00 ft = 3048 cm, and h = 6 ft, one finds,

0
s − f − 15
m=− = =
s s 3048
1 5
0
h = | m| h = 6 ft = 0. 0295 ft = 0. 9 cm
3048
( b) Distant objects form an image in the first lens at f1 = 1 2 cm. Thus, the object distance for lens 2 is
s 2 = ( 8 − 1 2) cm = − 4 cm . Then the image distance from lens 2 is s 20 = 1 5 cm so the focal length of the
second lens must be,
1 1 1 1 1
= + = + ⇒ f2 = − 5. 45 cm
f2 s 2 s 20 − 4 cm 1 5 cm
Then for the person at 1 00 feet,
0
! 0
!
s s − 12 − 15
m = m1 m2 = − 1 − 2 = − = − 0. 01 48
s1 s2 3048 −4
The ratio of the magnifications is

| m| 1 5/ 3048 1
= =
| m ( telephoto) | 1 80/ 4 ( 3048) 3
18
f 50 mm
3-2 0. ( a) D = = = 27. 8 mm
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 = 1 6. 0 mm; D = 50/ 5. 4 = 9. 26 mm; D = 50/ 9. 4 = 5. 35 mm
1
( d) f/ 1 . 8 at is equivalent in exposure to f/ 3. 1 at 3/ 1 00 s, f/ 5. 4 at 9 / 1 00 s, f/ 9. 4 at 27/ 1 00 s
1 00
25 1 1 1 L
3-2 1 . 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

1 1 1 f + f2 1 1 1 25 1 1
= + − 1 = + and M = = 1 2. 5 +
feq f1 f2 2 f1 f2 2 f1 f2 feq f1 f2
3-2 2 . The focal lengths of the two lenses are, f1 = f2 = 3 cm and they are separated by L = 2. 8 cm.
( a) The equivalent focal length is,
1 1 1 L 1 1 2. 8
= + − = + − feq = 2. 81 25 cm
feq f1 f2 f1 f2 3 3 9
( b) The magnification is,
25 25
M= +1= + 1 = 9. 9 ≈ 1 0 ×
feq 2. 81 25
3-2 3. ( a) The system is sketched below.

Ob j Ocular s0
Fo b j M = m ob j × Mo c = × 1 0 where s 0 = f + 1 6 = 1 6. 5 cm
0
s
s f ( 1 6. 5) ( 0. 5) 8. 25
s= 0 = =
1 6 cm s −f 1 6. 5 − 0. 5 16
1 6. 5 cm s
0
( 1 6. 5) ( 1 6)
| m| = = = 32, M = ( 32) ( 1 6) = 320 ×
s 8. 25
8. 25
( b) s = = 0. 51 6 cm
16
0
s ob j 25 s f ( 1 . 20) ( 1 )
+ 1 where, s ob j = 0 ob j =
0
3-2 4. ( a) M = m ob j × Mo c = = 6 cm. So,
s ob j fo c s 0 − fob j 1 . 20 − 1

6 25
M= + 1 = 46. 7 ×
1 . 20 3
0
s f ( − 25) ( 3)
( b) L = s ob j + s o c = 6 + 2. 68 = 8. 68 cm where s o c = 0 o c o c =
0
= 2. 68 cm
s o c − fo c − 25 − 3

25 L 25 25 − 4 − fo
3-2 5. M = − and L = d − fo − fe ⇒ − 20 = − ⇒ fo = 5 cm
fe fo 4 fo
0 0
1 1 1 0 s 0. 201 3 0 h 0. 001
3-2 6. + = ⇒ s = 0. 201 3 m, m = = = 0. 00671 1 , h = m h ⇒ h = = = 0. 1 49 m = 1 4. 9 cm
30 s 0 0. 2 s 30 m 0. 00671 1
19
fob j 14 D 35
3-2 7. ( a) M = 7 × ( b) M = or fo c = = 2 cm ( c) M = ob j or D ex = = 5 mm
fo c 7 D ex 7
0 sf L foc ( 1 4 + 2) ( 2) 16
( d) The exit pupil position is the image of L 1 in L 2 , or s = = = = = 2. 3 cm
s − f L − foc ( 1 4 + 2) − ( 2) 7
( e) Let y be the linear dimension of the field of view at a distance of x = 1 000 yd. Then the angular field of
view can be written as,
Df 1.8 y
θ= = = 0. 1 1 25 rad = ⇒ y = x θ = ( 1 000 yd) ( 0. 1 1 25 rad) = 337 ft.
L 14+ 2 x
The field of view is 337 ft at 1 000 yd. The geometry is sketched ( not to scale) below.
Obj
y θ θ Df
x L
3-2 8. ( 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 0 fob j
y = fob j tan α = s o c tan α 0 ⇒ =
tan α s oc
α 0 tan α 0 fob j
The angular magnification is then, M = ≈ =
α tan α s oc
Since m o c = − s / s o c , the required result follows,
00
fob j m o c
M=−
s 00
O b jec t ive O c u lar
fo b j s oc
α α0
α α0
O b jec t
s 00
F in al im ag e
( b) When viewed at ∞ , M = fo / fe = 30/ 4 = 7. 5 × . When viewed at 25 cm, s 00 = − 25 cm and so,

s 00 s 00 s 00 − fo c ( − 25) − 4 29
mo c = − = − 00 00
=− =− = = 7. 25 ×
s oc s fo c / ( s − fo c ) fo c 4 4
Them by the formula from part ( a) ,
( 30) ( 7. 25)
M=− = 8. 70 ×
( − 25)
3-2 9. The image size on the moon is determined by the angle α 0 shown in the figure associated with the solution
to problem 3 28 above. In that problem solution it is shown that M = α 0 / α. Here α = 0. 5 ◦ = π/ 360. The
diameter of the image of the moon is given by D M o on = s 00 α 0 = ( 25 cm) α 0 . To find α 0 use the relation
obtained in the solution to problem 3 28) , M = fob j / s o c . Here, from a rearrangement of the thin lens for
mula,
s 00 f ( − 25) ( 5 ) 25
s o c = 00 o c = cm = cm
s − fo c − 25 − 5 6
so that,
20 24
M= = ×
( 25/ 6 ) 5
Then,
24 π π π π
α0 = Mα = = and D M o on = s 00 α 0 = 25 cm = cm = 1 . 05 cm
5 360 75 75 3
20
3-30. ( a) L = fob j + fo c = 1 2 cm − 4 cm = 8 cm. M = fob j / fo c = ( 1 2/ 4) = 3 × .
( b) For the image at s 00 = 30 cm:
s 00 f − 30 ( − 4) 60
s o c = 00 o c = cm = − cm
s − fo c − 30 + 4 13
f 12
M = ob j = = 2. 6 ×
s oc 60/ 1 3
L = fob j + s o c = 1 2 cm − ( 60/ 1 3) cm = 7. 38 cm
3-31 . See Figure 3 39 in the text. In the normal position, s o c = fo c = 5 cm. But when s 00 = 25 cm,
s 00 fo c 25 · 5
s oc = = cm = 6. 25 cm
s 00 − 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∞ = = = 1 2. 5 × ( b) Mt el = 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 0 = 1 / 2, s 0 → ∞ ) . Given this separation, the image of the objective in the field
lens is
30( 2) 15
s 10 = 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)
s 20 = cm = 0. 1 33 cm
− 1 /7 − 2
Thus the exit pupil falls 0. 1 33 cm from the eye lens.
4. 5
Using M = D ob j / D E x P we also have D E x P = cm = 3 mm.
15
( d) θ = D F L / L = 2/ 30 = ( 1 / 1 5) 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,
α m h/ fe fo
| M| = = =
αo h/ fo fe
3-34. Refer to figure 3 41 in the text. For M2 :

1 1 1 s s0 ( − 2) ( 1 0)
+ 0= ⇒f= = ft = − 2. 5 ft
s s f s+s 0 − 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/ 1 2
The magnification is increased by the convex mirror by the factor, m = − s 0 / s = − ( − 1 0) / ( − 2) = + 5
The overall magnification is then, Moverall = m M = ( 5) ( − 36) = − 1 80 ×
21
C hapter 4 Wave Equations
4-1 . The waveform is sketched below. As a function of time y = A e − b ( x + 1 0 t)

2
4-2 . ( a) Replace x by x + v t:
4 m3
y( x , t) =
( x + ( 2. 5 m/ s) ) 2 + 2 m 3
( 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,

( 1 00 m 2 ) x − ( 1 0 m / s) t) 2 / m 2
y= e
x − ( 1 0 m/s) t
Thus y = y( x − ( 1 0 m/ s) t) and so the waveform is a traveling wave with v x = 1 0 m/ s.
4-5. From the given information, k = 2 π/ λ = ( 2 π/ 5) m

−1
, | v | = f λ = λ/ T = ( 5/ 3 ) m/ s, v z = − 5/ 3 m/ s
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 e 2 π i ( z / λ ± t/ T) = 2 e 2 πi [ z / ( 5 m ) + t/ ( 3 s) ] , 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 ± v x t) = ( 5 m) sin ( x + ( 2 m/ s) ( 4 s) ] = ( 5 m) sin ( x + 8 m)
50 m 25 m
22

x t 628. 3 x 6283
4-7. Compare the general form, y = sin 2 π − to the given form, y = ( 1 0 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 = 1 000 Hz
( c) k = 2 π/ λ = 628. 3/ cm ( d) ω = 2 πν = 2 π( 1 000) / s = 6283/ s ( e) T = 0. 001 s
( f) v = ν λ = ( 1 000) ( 0. 01 ) cm/ s = 1 0 cm/ s ( g) A = 1 0 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) ϕ 2 = B z 2 + B C 2 t 2 − 2 B C z t = B ( z − C t) 2 , ϕ = B ( z − C t) = constant, =0= B −C .
dt dt
dz
Therefore, v z = = 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) = 1 3 = 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,
− 1 3/ 7. 5 = − tan ϕ ⇒ ϕ = tan − 1 ( 1 3/ 7. 5) = 1 . 047 ≈ π/ 3
Then, from ( 1 )
A = 1 3/ sin ϕ = 1 3/ sin( π/ 3 ) = 1 5. 0
So that,
y = 1 5 sin( 2 π x/ λ + π/ 3 )
4-1 0. ( a) Take the waveform at t = 0 to be y = A sin( 2 π x/ λ + ϕ 0 ) . We are given that y = A at x = x 0 . That is,
y( x 0 ) = A sin( 2 π x 0 / λ + ϕ 0 ) = A
sin( 2 π x 0 / λ + ϕ 0 ) = 1
2 π x 0 / λ + ϕ 0 = π/ 2
π 2π
ϕ0 = − x0
2 λ
( b) For λ = 1 0 cm,
x 0 = 0, ϕ 0 = π/ 2
π
x 0 = 5/ 6 cm, ϕ 0 =
3
x 0 = 5 / 2 cm, ϕ 0 = 0
π
x 0 = 5 cm, ϕ 0 = −
2
3π
x 0 = − 1 / 2 cm, ϕ 0 =
5
( c) Generally y = A cos( 2 π x/ λ + ϕ c ) the same argument given in ( a) requires that
cos( 2 π x 0 / λ + ϕ c ) = 1
2 π x0 / λ + ϕc = 0
ϕ c = − 2 π x 0 / λ = ϕ 0 − π/ 2
Thus we should subtract π/ 2 from the answers in part ( b)
23
4-1 1 . Generally, k · r = k x x + k y y + k z z.
( a) For propagation along the z axis, k x = k y = 0, So k · r = k z z, with k z = 2 π/ λ. The waveform can then be
written as,

2π
ψ = A sin( k · r − ω t) = A sin( k z z − ω t) = A sin ( z − v t) = A sin 2 π( z/ λ − v t)
λ
√ 1 2π
( b) In this case, k z = 0 and k x = 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( k x 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
4-1 2 . 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 = e i θ = cos θ + i sin θ and apply the result from ( a) : cos θ = e i θ + e − i θ / 2

( d) Let z̃ = e i θ = cos θ + i sin θ and apply the result from ( b) : sin θ= e i θ − e − i θ / 2i
4-1 3. ( a) Note that e i π/ 2 = cos( π/ 2) + i sin( π/ 2) = i so that
i A e i ( k · r − ω t + ϕ 0 ) = e i π/ 2 A e i ( k · r − ω t + ϕ 0 ) = A e i ( k · r − ω t + ϕ 0 + π/ 2 )
( b) Similarly, e i π = cos( π) + i sin( π) = − 1 so that
− A e i ( k · r − ω t + ϕ 0 ) = e i π A e i ( k · r − ω t + ϕ 0 ) = A e i ( k · r − ω t + ϕ 0 + π)
4-1 4. One can write ψ = Im( ψ̃ ) with
ψ̃ = A e i ( k y + ω t ) + A e i ( k y − ω t + π) = A e i ( k y + ω t) + A e i π e i ( k y − ω t) = A e i ( k y + ω t) − A e i ( k y − ω t)
where I have used the result from problem 4 1 3b. Regrouping gives,

ψ̃ = A e i k y e i ω t − e − i ω t = 2 i A e i k y sin( ω t) = 2 A sin( ω t) [ i cos( k y) − sin( k y) ]
Here I have used the result from problem 4 1 2d. Reforming the real wavefunction gives the desired relation:
ψ = Im( ψ̃ ) = 2 A cos( k y) sin( ω t)
4-1 5. The irradiance is related to the electric field amplitude by the relation Ee = ( ε 0 c/ 2) E02 so that,
r r
2 Ee 2000
E0 = = V/ m = 870 V/ m
ε0 c 8. 85 × 1 0 − 1 2 × 3 × 1 0 8
Then,
B0 = E0 / c = 870/ ( 3 × 1 0 8 ) T = 2. 90 × 1 0 − 6 T
24
4-1 6. ( a) B0 = E0 / v = E0 n/ c = ( 1 00) ( 1 . 5) / ( 3 × 1 0 8 ) T = 5. 0 × 1 0 − 7 T
( b) h S i = Ee = ε 0 c 2 / 2) E0 B0 = ( 8. 85 × 1 0 − 1 2 9. 0 × 1 0 1 6 / 2) ( 1 00) ( 5. 0 × 1 0 − 7 ) W/m 2 = 1 9. 9 W/ m 2
4-1 7. ( a) From the solution to problem 4 1 5,

r r
2 Ee 2 · 1 350
E0 = = V/ m = 1 01 0 V/m, B0 = E0 / c = 3. 37 × 1 0 − 6 T
ε0 c 8. 85 × 1 0 − 1 2 × 3 × 1 0 8
Ee Ee 1 350
( b) Photon flux = = = 3 4
m − 2 s − 1 = 4. 76 × 1 0 2 1 m − 2 s − 1
h ν h c/ λ ( 6. 626 × 1 0 − ) ( 3 × 1 0 8 ) / ( 7 × 1 0 − 7)
( c) k = 2 π/ λ = 2 π/ ( 7 × 1 0 − 7 m) = 1 . 43 × 1 0 6 / m, ν = c/ λ = 4. 28 × 1 0 1 4 Hz. So,

E = ( 1 01 0 V/m) sin 2 π 1 . 43 × 1 0 6 / m s ± 4. 28 × 1 0 1 4 / s t
220 · 0. 05
4-1 8. ( a) Ee = Φ / A = W/ m 2 = 8. 75 × 1 0 − 3 W/m 2 ,
4 π ( 1 0) 2
r r
2 Ee 2 · 8. 75 × 1 0 − 3
E0 = = V/ m = 2. 57 V/ m
ε0 c 8. 85 × 1 0 − 1 2 × 3 × 1 0 8
2000
( b) Ee = Φ / A = − 1 0 W/ m 2 = 2 × 1 0 1 3 W/m 2
r r0
1
2 Ee 2 · 2 × 1 01 3
E0 = = V/ m = 1 . 23 × 1 0 8 V/ m, B 0 = E0 / c = 0. 409 T
ε0 c 8. 85 × 1 0 − 1 2 × 3 × 1 0 8
4-1 9. The energy flow in a cylindrical wave is in the ( cylindrical) radial direction. The total flux Φ e 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,
Φ e = 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 p
Ee ∝ ∝ E02 ⇒ E0 ∝ 1 / r
r
4-2 0. Eq. ( 4 44) gives,
λ 0 ( 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 / c 2 ) ≈ 1 − v/ c
Using this in the expression for λ 0 / λ gives the desired result,
λ0 p
≈ ( 1 − v/ c) ( 1 − v/ c) = 1 − v/ c
λ
r
λ 0 540 1 − v/ c
4-2 1 . = = ⇒ v = 0. 1 68 c
λ 640 1 + v/ c
r
λ0 1 − v/ c
4-2 2 . = 4. 8 = ⇒ v = − 0. 91 7 c
λ 1 + v/ c
25
4-2 3. From thermodynamics:
r r
3RT 3 ( 8. 31 ) ( 1 000)
v rm s = = = 2497 m/s
M 0. 004
Since the speed is small compared to c, we can use Eq. ( 4 45) ,
λ0
= 1 − v/ c ⇒ ( λ 0 − λ) = ∆λ = ( v/ c) λ
λ
2497
∆λ = 706. 52 × 1 0 − 9 m = 0. 0059 nm
3 × 10 8
Now ∆ λ is added to λ 0 = λ( v = 0) for receding atoms and subtracted from λ 0 for approaching atoms. Thus
an estimate for the Doppler width is
Doppler Width ≈ 2 ∆λ = 0. 01 2 nm
4-2 4. 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
C hapter 5 S uperposition of Waves
5-1 . ( a) E1 is a function of u 1 = k x − ω t = ( 3 / m) x − ( 4/ s) t. The velocity component of the wave can be deter

mined by setting u 1 = constant so that,
du 1 dx dx
=0=k − ω ⇒ vx = = ω/ k = ( 4/ 3 ) m/ s
dt dt dt
Similarly, E2 is a function of u 2 = k x + ω t = ( 3 / m) x + ( 4/ s) t and so,
du 2 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,
5 E0 5 E0
E1 + E 2 = 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 − i ω t , E2 = 7 e i π/ 4 e − i ω t
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 = E02 1 + E022 + 2 E0 1 E0 2 cos( α 2 − α 1 )
p
E0 = 2 2 + 7 2 + 2 · 2 · 7 cos( π/ 4) = 8. 53
E Using Eq. ( 5 1 0) :
√
E2 E0 1 sin( α 1 ) + E0 2 sin( α 2 ) 0 + 7/ 2
7 tan( α) = = √ ⇒ α = 0. 61 9in ≈ 0. 2 π
E0 1 cos( α 1 ) + E0 2 cos( α 2 ) 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 1 0) ,
p p
E0 = E021 + E022 + 2 E0 1 E0 2 cos( α 2 − α 1 ) = 3 2 + 4 2 + 2 · 3 · 4 cos( π/ 2 − π/ 6) = 6. 08
E sin( α 1 ) + E0 2 sin( α 2 ) 3 sin( π/ 6) + 4 sin( π/ 2)
tan α = 0 1 = ⇒ α = 0. 36 π
E0 1 cos( α 1 ) + E0 2 cos( α 2 ) 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 1 0) . 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 1 0) :
p
y0 = 5 2 + 7 2 + 2 · 5 · 7 cos( π/ 6 ) = 1 1 . 6
0 + 7 sin( π/ 6 )
tan α = = α = 0. 098 π
5 + 7 cos( π/ 6)
y = 1 1 . 6 cos( 0. 098 π − ω t) = 1 1 . 6 cos( ω t − 0. 098 π) = 1 1 . 6 sin( ω t − 0. 098 π + π/ 2)
y = 1 1 . 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/ 1 8 − 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 1 3) 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 1 2) ,
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 = ( 1 00) ( 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
ER2 = ( 1 00) E02 1 ⇒ ER = 1 0 ( 0. 02 V/ m) = 0. 2 V/ m
5-7. At the indicated position,

ψ 1 = A 1 cos( 8 π/ 3 − ω t) , A 1 = 4 cm, ω = 20/ s
ψ 2 = A 2 cos( 3 π/ 2 − ω t) , A 2 = 2 cm, ω = 20/ s
Using Eqs ( 5 9) and ( 5 1 0) ,
p p
A = A 21 + A 22 + 2 A 1 A 2 cos( α 2 − α 1 ) = 20 + 1 6 cos( 3 π/ 2 − 8 π/ 3 ) cm = 2. 48 cm
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)
ψ R = ( 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.

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

2 π c ω 2 dn ω dn
vg = vp − = v p 1 −
ω 2 π n 2 dω n dω
( b) For normal dispersion, dn/ d λ < 0 so that dn/ dω > 0 and v g > v p.
28
nD − nC 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 = n C − n F = ∆ ( n 0 − 1 ) = ( 1 . 50 − 1 ) = 1 / 60 and ∆ λ = λ C − λ F = ( 656. 3 − 486. 1 ) nm = 1 70. 2 nm
30
Then,

c λD ∆n c 589 1
vg = 1+ = 1− = c/ 1 . 56
nD nD ∆ λ 1 . 50 1 . 5 60 · 1 70. 2
5-1 0. Now, n = A + B/ λ 2 and v g = v p [ 1 + ( λ/ n) dn/ dλ] . Therefore,

dn 2 ( 2. 5 × 1 0 6 ) ˚ − 1 −1
= − 2 B/ λ 3 = − A = − 4 × 1 0 − 5 ˚A
dλ ( 5000) 3
Then,

c c c 5000 c
vp = = and v g = 1+ − 4× 10 −5
=
n 1 . 50 1 . 50 1.5 1 . 73
√
5-1 1 . ( a) We know that K = n 2 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 K1 / 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 − ω 2 where the second term is small compared to the first. Then,
dK − A ( − 2 ω) 2Aω
= 2
=
dω 2
ω0 − ω ) 2 ω02 − ω 2 ) 2
Using the result of ( a) , ! !
c ω 2Aω c A ω2
vg = √ 1− =√ 1−
K 2 K ω02 − ω 2 ) 2 K 2
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 ω02 − ω 2 ω02 − 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 −
ω02 − ω 2 ) 2
29
dω dv
5-1 2 . ( a) According to Eq. 5 38, v g = = v p + k p . Then with k = 2 π/ λ and dk = − ( 2 π/ λ 2 ) dλ. Thus,
dk dk
2 π d vp λ 2 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
5-1 3. As in problem 5 1 2a, v g = v p − λ . Then,
dλ
Long λ:
q q
dv p d g 1 g 1
= = = vp
dλ dλ 2 πλ 2 2 πλ 2
Short λ:
1/2 1 /2
dv p d 2 πT 1 2 πT
= =−
dλ dλ λρ 2λ λρ
So,
1 /2 1/2 1 /2
2 πT 1 2 πT 3 2 πT 3
vg = + = = vp
λρ 2 λρ 2 λρ 2
5-1 4. The frequency received by the mirror is ν10 = ν0 / ( 1 + v/ c) . Since λ 0 / λ = ν/ ν 0 = 1 − v/ c and v < 0 due to
recession. The reflected beam is also due to a receding source so the reflected frequency is
ν10
ν20 =
1 + v/ c
Thus,
ν0 1 ν0
ν20 = = = 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 ν20 ≈ ν0 ( 1 − 2v/ c) and the beat frequency is
ν20 − ν0 = 2( v/ c) ν0
5-1 5. 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 = ( 1 4 cm) sin[ ( 4/ m) x] cos[ ( 2 π/ T) t]
This wave has an amplitude of 1 4 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
5-1 6. The standing wave has the form,
πx
50 π
y = 2 A sin( k x) cos( ω t) = ( 3 cm) sin cos t
1 0 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/ 1 0) ) cm/ s = 500 cm/ s. The
wavelength is λ = 2 π/ k = ( 2 π) / ( π/ 1 0) cm = 20 m. The frequency is ν = ω/ 2 π = 25 Hz.
( b) The internodal distance is λ/ 2 = 1 0 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) = − ω 2 y
dt
a y ( x = 5 cm, t = 0. 22 s) = 7. 40 × 1 0 4 cm/s 2
5-1 7. In the complex representation,

h i
E1 = E0 sin( ω t + k x) = Im E0 e i ( ω t + k x )
h i
E2 = E0 sin( ω t − k x − ϕ R ) = Im E0 e i ( ω t − k x − ϕ R )
h i
ER = E1 + E2 = E0 Im e i ( ω t + k x ) + e i ( ω t − k x − ϕ R )
h i
ER = E0 Im e i ( ω t − ϕ R / 2 ) e i ( k x + ϕ R / 2 ) + e i ( ω t − ϕ R / 2 ) e − i ( k x + ϕ R / 2 )
n h io
ER = E0 Im e i ( ω t − ϕ R / 2 ) e i ( k x + ϕ R / 2 ) + e − i ( k x + ϕ R / 2 )
h i
ER = E0 Im e i ( ω t − ϕ R / 2 ) 2 cos( k x + ϕ R / 2)
h i
ER = 2 E0 cos( k x + ϕ R / 2) Im e i ( ω t − ϕ R / 2 )
ER = 2 E0 cos( k x + ϕ R / 2) sin( ω t − ϕ R / 2)
5-1 8. The separation between cavity modes is ∆ ν = νm + 1 − νm = ( m + 1 ) c/ 2d − m c/ 2d = c/ 2d. That is,

∆ ν = 3 × 1 0 8 m/s / ( 2 · 1 m = 0. 1 5 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. 1 5 GHz
31
C hapter 6 P roperties of Lasers
6-1 . The allowed energies of the electron in the hydrogen atom are En = − ( 1 3. 6 eV) / n 2 , n = 1 , 2, 3 , . . . The
energies required for this problem are:
E1 = − 1 3. 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 1 240 eV · nm
λmn = =
Em − En Em − En
( a) Lyman Series:
1 240 1 240
λ21 = nm = 1 22 nm λ31 = nm = 1 03 nm λ 4 1 = 97. 3 nm
− 3. 4 − ( − 1 3. 6) − 1 . 51 − ( − 1 3. 6)
These are in the ultraviolet ( shorter than visible) .
( b) Balmer Series:
1 240 1 240
λ32 = nm = 656 nm λ 42 = nm = 486 nm λ 5 2 = 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 Efi n al − Ein it ial = 0 − ( − 1 3. 6 eV) = 1 3. 6 eV. Any energy greater than this is possible
since the free electron can have any energy. The corresponding wavelength range is,
hc 1 240 eV · nm
λ< = = 91 . 2 nm
1 3. 6 eV 1 3. 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 1 240 eV · nm
λ< = = 365 nm
3. 4eV 3. 4 eV
6-3. ( a) The rotational inertia is

I = 2 m H r 2 = 2( 1 . 67 × 1 0 − 2 7 0. 074 × 1 0 − 9 / 2) 2 kg · m 2 = 4. 6 × 1 0 − 4 8 kg · m 2
( b) ∆ E = E1rot − E0rot = ~ 2 / I = ( 6. 626 × 1 0 − 3 4 / 2 π) 2 / ( 4. 6 × 1 0 − 4 8 ) J = 2. 4 × 1 0 − 2 1 J = 0. 01 5 eV
( c) P1 / P0 = e − ( ∆ E ) / k B T = e − 0 . 0 1 5 / 8. 62 × 1 0 − 5 · 293)
= 0. 55

6-4. ( a) ∆E = E1v ib − E0v ib = h f = ( 6. 626 × 1 0 − 3 4 ) 1 . 3 × 1 0 1 4 J = 8. 6 × 1 0 − 2 0 J = 0. 54 eV
( b) P1 / P0 = P1 / P0 = e − ( ∆ E ) / k B T = e − 0 . 5 4 / 8. 62 × 1 0 − 5 · 293)
= e− 21 . 4 ≈ 5 × 1 0− 1 0
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. 01 5 eV
E2 , 0
E2 , 0 = E0 , 0 + 1 . 08 eV
E1 , 0 + 0 . 0 6 eV
E1 , 0 + 0 . 0 1 5 eV
E1 , 0 = E0 , 0 + 0. 54 eV
E1 , 0
E0 , 0
E0 , 0 + 0. 06 eV
E1 , 0 + 0. 01 5 eV
E1 , 0
6-6. The energy separation of the states is ∆E = h c/ λ = ( 1 240 eV · nm) / ( 632. 8 nm) = 01 . 96 eV
Neglecting degeneracies: N2 / N1 = P2 / P1 = e − ( E 2 − E 1 ) / k B T = e − 1 . 9 6 / 8. 62 × 1 0 − 5 · 300)
≈ e − 76.
6-7. The differential equation can be integrated as shown below,

dN2 dN2
= − A 2 1 N2 ⇒ = − A 2 1 dt
dt N2
Z N2 ( t) Zt
dN2
= − A2 1 dt 0 ⇒ ln( N2 ( t) / N2 0 ) = − A 2 1 ( t − 0) = − A 2 1 t
N2 0 N2 0
N2 ( t)
= e − A 2 1 t ⇒ N2 ( t) = N2 0 e − A 2 1 t
N2 0
At time t = 1 / A 2 1 , N2 ( t) = N2 0 e − A 2 1 ( 1 / A 2 1 ) = N2 0 e − 1 = N2 0 / e
2 π h c2 1
6-8. The black body spectral radiance is Mλ = .
λ 5 eh c/ λ kBT − 1
This function is maximized when it’ s denominator is minimized:

d h 5 h c/ λ k B T i
5 h c/ λ k B T h c 1 h c/ λ kBT

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

with solution, x = 4. 999986. Then,
hc hc
x m ax = 4. 999986 = ⇒ λ m ax T = = 2. 898 × 1 0 − 3 m · K = 2898 µm · K
λ m ax k B T 4. 999986 k B

6-9. Let x = h c/ ( λ k B T) . Then dx = ( h c/ k B T) − dλ/ λ 2 so that, dλ = − λ 2 ( k B T/ h c) dx. Then,
Z∞ Z∞ Z∞
2 π h c2 1 2 π ( k B T) 4 x3 2 π ( k B T) 4 π 4 2 kB4 5
π
M= Mλ dλ = dλ = = = T4
0 0 λ5 e h c/ λ kBT − 1 h 3 c2 0 ex − 1 h 3 c2 15 1 5 h 3 c2

2 kB4 5
π
M= T 4 ≡ σ T4 = 5. 668 × 1 0 − 8 W/ m 2 K 4
1 5 h 3 c2
33
6-1 0. 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,

hc dλ λ2 kB T
dx = − 2 ⇒ dλ = − dx
kB T λ hc
So,
Z0 Z∞ Z∞ 3
1 1 − λ 2 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 k B T
∞ λ e − 1 hc 0 λ3 ex − 1 0 hc ex − 1
Z
2 π k B T) 4 ∞ x 3
M= dx
h 3 c2 0 ex − 1
The definite integral has the value:
Z∞
x3 π4
dx =
0 ex −1 15
so that,
4 5
4
2k B π W
M= T4 ≡ σ T4 , σ = 5. 668 × 1 0 − 8
1 5 h 3 c2 m2 K4
6-1 1 . The original temperature is determined from the Wien displacement law,
2898 µm · K 2898 µm · K
λ m a x T1 = 2898 µm · K ⇒ T1 = = = 5269 K
λ m ax 0. 55 µm
For the total radiant exitance to double,
M2 = 2 M1 ⇒ σ( T2 ) 4 = 2σ T14 ⇒ T2 = 2 1 / 4 T1 = 6266 K
The new maximum wavelength is,
2898 µm · K 2898 µm · K
λ m ax = = = 0. 462 µm = 462 nm
T2 6266 K
6-1 2 . The gray body at T2 must match the the black body at the lower temperature, T1 . That is,

( MG B ) T2 = ( MB B ) T1 ⇒ ( εMB B = ( MB B ) T1
T2
ε σ T24 = σ T14 ⇒ T24 = T14 / ε = ( 5000 K) 4 / 0. 45 ⇒ T2 = 61 05 K
2 π h c2 1
6-1 3. On the absolute temperature scale 451 ◦ F = 506 K. Then Mλ = ε 5 h c/ λ k
. 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
6-1 4. Equation ( 6 1 2) indicates that A 2 1 / B2 1 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.
6-1 5. The frequency bandwidth ∆ν = ν2 − ν1 is related to the wavelength range ∆λ = λ 2 − λ 1 by,
c c λ − λ2 ∆λ λ 2 | ∆ν |
∆ν = − =c 1 ≈ − c 2 ⇒ | ∆λ | =
λ2 λ1 λ2 λ1 λ c
2
1 . 06 × 1 0 − 6 1 . 2 · 1 01 1
From Table 6 1 , λ = 1 . 06 × 1 0 − 6 m, so, | ∆ λ | = m = 4. 5 × 1 0 − 1 0 m = 0. 45 nm
( 3 · 1 08 )

6-1 6. ( a) t c ≈ 1 / ∆ ν = 1 / ( 1 0 5 Hz) = 1 0 − 5 s L c = c t c = 3 · 1 0 8 m/ s 1 0 − 5 s = 3000 m

( b) t c ≈ 1 / ∆ν = 1 / ( 2 · 1 0 9 Hz) = 5 · 1 0 − 1 0 s L c = c t c = 3 · 1 0 8 m/ s 1 0 − 5 s = 0. 1 5 m = 1 5 cm
6-1 7. The separation between cavity modes is ∆ ν = νm + 1 − νm = ( m + 1 ) c/ 2d − m c/ 2d = c/ 2d. That is,

∆ν = 3 × 1 0 8 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
6-1 8. ( a) The half angle beam spread is,
λ 6. 33 · 1 0 − 7 m
θ= = = 4. 0 × 1 0 − 4 rad = 0. 023 ◦
π w0 π · 5 · 1 0 − 4 m
( b) The diameter is about,

D = 2 d θ = 2 ( 1 000 m) 4. 0 × 1 0 − 4 = 0. 8 m = 80 cm
6-1 9. Assuming that that irradiance is constant over the spot:
P P
Ilaser = laser = las2 er = Isu n
A la ser πD / 4
r r
4 Plas er 4 · 0. 01
D= = m = 0. 0036 m = 3. 6 mm
π Isu n π 1 000
35
6-2 0. ( a) The energy difference ∆ E between the levels must have near the same energy as the pump photons.
That is,
hc hc 1 240 eV · nm
⇒λ=
∆E = h ν = =
λ ∆E ∆E
Then, λ 1 . 5 3 = ( 1 240/ 1 . 53) nm = 81 0 nm. Similarly, λ 1 . 6 5 3 = 750 nm, λ 2 . 1 1 9 = 585 nm, and λ 2 . 3 6 1 = 525 nm.
( b) The quantum efficiency η Q is,
Energy of an output photon h c/ λ ou t λ
ηQ = = = p
Energy of a pump event h c/ λ p λ ou t
So using the results from part ( a) for the various transitions:
81 0 750 585 525
ηQ , 1 . 53 = = 0. 76, η Q , 1 . 6 5 3 = = 0. 70, η Q , 2 . 1 1 9 = = 0. 55, and η Q , 2 . 3 6 1 = = 0. 49.
1 064 1 064 1 064 1 064
6-2 1 . ( a) The laser output beam would have a power of Pou t p u t = 0. 8 × 0. 3 × 0. 7 × 0. 1 5 × 0. 5 × 2500 W = 31 . 5 W.
Pou t p u t 31 . 5
( b) The overall efficiency is η = = = 0. 01 26.
Ppu m p 2500
6-2 2 . The diode laser beam is generated in a very small region. The radius w 0 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
1 6) , the divergence angle is inversely proportional proportional to w 0 and so will be larger for diode lasers
than for other types of lasers.
6-2 3. 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 Nu p p er > Nlow er . The magnitude of the net rate of depletion of the population inversion due to
stimulated processes is proportional to I ( Nu pp er − Nlow er ) . Thus increased irradiance causes a larger rate of
depletion of the population inversion, leading to a lessened inversion.
6-2 4. 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
C hapter 7 Interference of Light

7-1 . ( a) I1 = ( ε 0 c/ 2) E02 1 = 8. 85 × 1 0 − 1 2 · 3 × 1 0 8 / 2 3000 ) 2 W/ m 2 = ( 0. 001 3275) 3000) 2 = 1 1 , 950 W/m 2
I2 = ( ε 0 c/ 2) E02 2 = ( 0. 001 3275) 4000) 2 = 21 , 240 W/ m 2
√ √
( b) I1 2 = 2 I1 I2 cos δ = 2 I1 I2 cos [ π/ 3 − ( π/ 5 − π/ 6 ) ] = 1 8, 720 W/ m 2
( c) I = I1 + I2 + I1 2 = 51 , 900 W/ m 2
( d) The visibility is,

√ √ √ √
Im ax − Im in I1 + I2 + 2 I1 I2 − ( I1 + I2 − 2 I1 I2 ) 4 I1 I2 2 I1 I2
visibility = = √ √ = = = 0. 96
Im ax + Im in 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 0 1 021 = = 0. 86
E 0 1 + E0 2 1 . 62 + 1 . 82
For perpendicular fields, I1 2 ∝ 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 0 1 022
I1 + I2 E 0 1 + E0 2
Then,
2 E0 1 E0 2 1 / E022 2R
V= 2 2 2 = R 2 + 1 , with R = E0 1 / E0 2
E0 1 + E 0 2 1 / E 0 2
For R = 2,
V = 0. 8
If V = 0. 5,
V ( 1 + R 2 ) = 2 R ⇒ 0. 5 + 0. 5 R 2 = 2 R
√
4± 16 − 4 √
R2 − 4R+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 = 1 3. 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 × 1 0 − 7 · 1 00
∆y = cm = 0. 0273 cm
0. 2
The irradiance on the screen relative to y = 0 at a maximum is given by,
h i
2 πay π 0. 2
I = 4 I0 cos = 4 I0 cos 2
y = 4 I0 cos 2 [ ( 1 1 5/ cm) y]
λL 546. 1 × 1 0 − 7 · 1 00
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 sin 2 [ ( 1 1 5/ cm) y]
7-6. Let λ 1 = 436 nm and the unknown wavelength be λ 2 . At the position of overlap,
( 3 + 1 / 2) λ 1 = a sin θ = 3 λ 2 ⇒ λ 2 = ( 3. 5/ 3 ) λ 1 = ( 7/ 6 ) ( 436 nm) = 508. 7 nm
7-7. See Figure 7 25 in the text. The position of the m t h 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. 1 5 × 1 0 − 5 cm = 51 4 nm
9L 9 ( 1 50)
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 × 1 0 − 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 × 1 0− 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 1 0− 4
∆ 2 − ∆ 1 = ∆m λ ⇒ ∆ m = = (n − 1) = (1.5 − 1) = 83. 3 fringes
λ λ 6 × 1 0− 7
π∆
( c) I = 4 I0 cos 2 where ∆ = a y/ L. At ∆ = 0, I = Im ax = 4 I0 . Then for I = 2 I0 = Im ax / 2:
λ
π∆
2 I0 = I = 4 I0 cos 2 ⇒ ∆ = λ/ 4 = 1 50 nm
λ
7-1 0. 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) 1 500 m + 1 /2
In the visible range, m = 4 and m = 5 give the wavelengths 556 nm and 455 nm.
38
7-1 1 . 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 × 1 0 − 9 1
α= = = 0. 005893 rad = 0. 3376 ◦ = 20. 3 0
2 n − 1 ∆λ 2 1.5 − 1 3 × 1 0− 4
7-1 2 . 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 × 1 0 − 7 ) ( 50 + 1 00)
θ= = = 0. 001 768 rad = 6 0 5 00
2 d∆ y 2 · 50 · 0. 05
7-1 3. 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/ 1 9. Then,
546. 1 × 1 0 − 7 ) ( d + 4 d)
α= = 0. 01 0376 rad = 35 0 40 00
2 d ( 1 . 5 − 1 ) 0. 5/ 1 9
7-1 4. 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 = ( m 1 + 1 / 2) λ 1 = ( m 2 + 1 / 2) λ 2
m 1 + 1 / 2 λ 2 675
= = = 1 . 2875
m 2 + 1 / 2 λ 1 525
By trial and error, this relation is satisfied with m 1 = 4 and m 2 = 3 . Then,
( m 1 + 1 / 2) λ 1 ( 4. 5) ( 525 nm)
t= = = 908. 65 nm
2n 2 ( 1 . 30)
7-1 5. At normal incidence, ( m + 1 / 2) λ = 2 n t. At 45 ◦ , ( m + 1 / 2) λ 0 = 2 n t cos θ t .

Here θ t 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 θ t ⇒ sin θ t = √ =√ = 0. 51 24 ⇒ θ t = 30. 825 ◦ = cos θ t = 0. 859
2n 2 ( 1 . 38)
Then,
2 n t cos θ t ( m + 1 / 2) λ cos θ t
λ0 = = = λ cos θ 1 = ( 580 nm) ( 0. 859) = 498 nm
m + 1 /2 m + 1 /2
√ p
7-1 6. Using Eq. ( 7 28) , the index of refraction of the film is n f = n 0 n s = ( 1 ) ( 1 . 78) = 1 . 33.
The required thickness is t = λ f / 4 = λ/ ( 4 n f ) = ( 550 nm) / ( 4 · 1 . 33) = 1 03 nm
39
1−n 1 − 1.4
7-1 7. ( a) r = where n = n 2 / n 1 . 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 = λ n f / 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 ot t om = = − 1 /1 5
1 + 1 . 6/ 1 . 4
An estimate gives
2 2
t ive = ( | r t op | − | r b ot t om | ) = ( 1 / 6 − 1 / 1 5) = 0. 01 ⇒ 1 % energy reflected
2
r effec
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.
7-1 8. 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 = 1 5,
θ
m = 15 ( 1 5) λ = 2 ( 1 . 33) t = 2. 66 ( 1 cm) θ ⇒ θ ≈ 1 5 λ/ 2. 66 = 3. 57 × 1 0 − 4 rad = 1 0 1 4 00

t
7-1 9. 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 θ t
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 · ( 1 0 4 nm) 0. 707 1 4, 1 42 nm
λm = = =
m m m
There are 1 5 orders in the visible with m ranging from 21 to 35. The dark lines occur at:
λ 2 1 = 673. 4 nm, λ 2 2 = 642. 8 nm. . . , λ 3 5 = 404. 1 nm.
7-2 0. Refer to Figure 7 1 5b 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 t h bright fringe corresponds
to m = 39 since the first bright fringe occurs for m = 0. Then, for the 40 t h fringe,
( 39. 5) ( 589 × 1 0 − 7 )
39. 5 λ = 2 t ⇒ t = cm = 1 . 1 6 × 1 0 − 3 cm
2
7-2 1 . 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 × 1 0 − 9 m)
For ∆m = 1 , ∆x = ( 0. 1 m) λ/ d = = 1 . 09 mm
d
5 × 1 0− 5 m
xd ( 0. 2 m) ( 5 × 1 0 m)
− 5
x
Also, solving for m = = = 1 81 . 3
2 0 cm ( 0. 1 m) λ ( 0. 1 m) ( 546. 1 × 1 0 − 9 m)
Counting the m = 0 dark fringe, 1 84 fringes would appear.
40
7-2 2 . The physical arrangement is sketched below.
The two image positions viewed by the reflected light are
I1 and I2000 . These serve as the virtual sources. There position
I10
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 I2000 is found as:
Reflection at top surface forming I20 :
s 1 1
+ = 0 ⇒ s 20 = − n s
I1
s s 20
Reflection at bottom forming I200 :
n
2 t/ n
s 2 = t − s 20 = t + n s ⇒ s 200 = − s 2 = − ( t + n s )
I20 0 0
Refraction out of top surface forming I2000 :
s 2 = t − s 200 = 2 t + n s

n 1 s 2t
+ = 0 ⇒ s 2000 = − 2 = − s+
s 2 s 2000 n n
I20 0
That is I2000 is formed a distance of s + 2 t/ n below the top
surface.
Thus the separation of virtual sources is 2 t/ n.
7-2 3. Refer to Figure 7 1 7 in the text. Using Eq. ( 7 38) ,

2 t m + ∆ r = 2 t m + λ/ 2 = m λ ⇒ t m = ( m − 1 / 2) λ/ 2
The 1 0 t h bright fringe occurs for m = 1 0, so that
t 1 0 = 9. 5 λ/ 2 = ( 9. 5 / 2) ( 546. 1 × 1 0 − 6 mm) = 2. 59 × 1 0 − 3 mm
Using Eq. ( 7 39) , the radius of curvature R of the lens surface can be found:
r12 0 + t 21 0 7. 89/ 2) 2 + 2. 59 × 1 0 − 3 ) 2
R= = = 3000 mm = 3 m
2 t1 0 2 ( 2. 59 × 1 0 − 3 )
7-2 4. Using Eq. ( 7 39) with t m rm ,

2
rm , air ≈ 2 R t m , air = 2 R ( m − 1 / 2) λ/ 2
2
rm , liq ≈ 2 R t m , liq = 2 R ( m − 1 / 2) λ/ 2n liq
2
rm , a ir r √
2 = n liq ⇒ m , air = n liq
r m , liq r m , liq
7-2 5. Using Eq. ( 7 39) with t m rm ,

r 12 1 , 1 ≈ 2 R t 1 1 , 1 = 2 R ( 1 1 − 1 / 2) λ 1 / 2 = r12 0 , 2 ≈ 2 R t 1 0 , 1 = 2 R ( 1 0 − 1 / 2) λ 2 / 2
λ 2 = λ 1 ( 1 0. 5/ 9. 5) = ( 546 nm) ( 1 0. 5/ 9. 5) = 603. 5 nm
p p
r1 1 = 2 R ( 1 1 − 1 / 2) λ 1 / 2 = 2 ( 1 ) ( 1 0. 5) ( 546 × 1 0 − 9 ) / 2 m = 0. 002398 m = 2. 39 mm
t 1 1 = 1 0. 5 λ 1 / 2 = 2. 87 × 1 0 − 4 cm
7-2 6. 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
7-2 7. 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 1 0− 3 W
( a) Ee = = 2= = 1 273 W/ m 2 = ( ε 0 c/ 2) E02 ⇒ E0 = 980 V/ m.
A πr π 5 × 1 0 − 4) 2 m 2
( b) Snell’ s law gives: ( 1 ) sin( 45 ◦ ) = ( 1 . 41 4) sin θ f ⇒ θ f = 30 ◦
( c) Using the Stokes relations,

| r 0 | = | r | = 0. 28
t t 0 = 1 − r 2 = 0. 921 6
( d) Reflected beams:
E1 = r E0 = ( 0. 28) 980 V/ m) = 274 V/m, ( E1 / E0 ) 2 = 0. 078 = 7. 8%
E2 = r 0 t t 0 E0 = ( 0. 28) ( 0. 921 6) ( 980 V/m) = 253 V/ m, ( E2 / E0 ) 2 = 0. 067 = 6. 7%
E3 = ( r 0 ) 3 t t 0 E0 = 0. 28) 3 ( 0. 921 6) ( 980 V/ m) = 1 9. 8 V/m, ( E3 / E0 ) 2 = 0. 00041 = 0. 041 %
( e) Transmitted beams:
E1 = t t 0 E0 = ( 0. 921 6) ( 980 V/m) = 274 V/ m, ( E1 / E0 ) 2 = 0. 85 = 85%
E2 = ( r 0 ) 2 t t 0 E0 = ( 0. 28) 2 ( 0. 921 6) ( 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 . 41 4) cos( 30 ◦ )
7-2 8. From Eq. ( 7 27) and the Stokes relations,

1 − n 1 − 1 . 52/ 1
0
| r| = | r | = = = 0. 206
1 + n 1 + 1 . 52/ 1
t t 0 = 1 − r 2 = 0. 95742
( a) Reflected beams:
E1 = r E0 = ( 0. 20635) ( 1 V/ m) = 0. 206 V/ m
E2 = r 0 t t 0 E0 = ( 0. 20635) ( 0. 95742 ) ( 1 V/ m) = 0. 1 98 V/m
E3 = ( r 0 ) 3 t t 0 E0 = ( 0. 20635) 3 ( 0. 95742) 1 V/ m) = 0. 084 V/ m
Transmitted beams:
E1 = t t 0 E0 = ( 0. 95742) ( 1 V/ m) = 0. 957 V/ m
E2 = ( r ) t t 0 E0 = ( 0. 20635) 2 ( 0. 95742) 1 V/ m) = 0. 041 V/ m
0 2
E3 = ( r 0 ) 4 t t 0 E0 = ( 0. 20635) 4 ( 0. 95742) 1 V/ m) = 0. 001 7 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. 1 98 = 1 . 044. V = = 0. 999
1 + 1 . 044 2
2 ( 23. 485)
Transmitted: R = E1 / E2 = 0. 957/ 0. 041 = 23. 485. V = = 0. 085
1 + 23. 485 2
42
C hapter 8 O ptical Interferometry
8-1 . λ = ( 2 ∆d/ ∆ m) = ( 2 · 0. 01 4 cm) / 523 = 4. 6 × 1 0 − 5 cm = 436 nm
8-2 . Straight fringes are due to a wedge between one mirror and the image of the other ( M2 and M1 0 in Figure
8 1 ) . Interference then occurs as from reflection by an air wedge.
546. 1 nm There are 1 2 fringes/cm so there are 1 1 fringe spaces/cm.
θ d m λ = 2 d ⇒ d = m λ/ 2 = ( 1 1 / 2) λ = 5. 5 λ.
1 cm θ = t/ ( 1 cm) = 5. 5 5. 461 × 1 0 − 6 ) / 1 = 3. 00 × 1 0 − 4 cm = 0. 01 72 ◦ = 1 0 2 00
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 × 1 0 − 9 m)
t= = = 23. 75 × 1 0 − 6 m = 23. 75 µm
2 (n − 1) 2 ( 1 . 434 − 1 )
8-4. ( a) Using Eq. ( 8 5) , m m a x = 2 d/ λ = ( 2 · 2 cm) / ( 500 × 1 0 − 7 cm) = 80, 000. ( b) m = m m 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 = = 1 53
589 × 1 0 − 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 θ m ⇒ cos θ 3 1 = = = 0. 98084 ⇒ θ 3 1 = 1 1 . 23 ◦
2d 20
The 1 0 t h dark ring is then of order m = 22:
22 λ 22 ( 0. 6328)
cos θ 2 2 = = = 0. 69608 ⇒ θ 2 2 = 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. 01 01 3 cm
2 2
43
8-9. ( a) The coefficient of finesse is,
4 r2 4 ( 0. 999)
F= = = 3. 996 × 1 0 6
( 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 × 1 0 − 9 ) ( 1 50 × 1 0 6 )
| ∆ν | = ∆λ ⇒ ∆ λ = | ∆ν | = m = 2. 00 × 1 0 − 1 3 m
λ 2 c 3 × 1 08
The resolving power required is then,
λ 632. 8 × 1 0 − 9
R= = = 3. 1 6 × 1 0 6
∆λ 2. 00 × 1 0 − 1 3
( c) The mode number is,
2R 2 ( 3. 1 6 × 1 0 6 )
m= √ = √ = 1 006
π F π 3. 996 × 1 0 6
The plate spacing is
m λ ( 1 006) ( 6. 328 × 1 0 − 5 cm)
d= = = 0. 031 8 cm
2 2
( d) The free spectral range is,
c 3 × 1 0 8 m/s
ν fs r = = = 4. 72 × 1 0 1 1 Hz
2 d 2 ( 3. 1 8 × 1 0 − 4 m)
Expressed as a wavelength range,
λ2 ( 632. 8 × 1 0 − 9 ) 2
λ fs r = ν fs r = 4. 72 × 1 0 1 1 m = 6. 300 × 1 0 − 1 0 m = 0. 63 nm
c 3 × 1 08
( e) The minimum resolvable wavelength difference is
λ2 λ2 λ2 ( 632. 8 × 1 0 − 9 ) 2
∆ λ m in = = √ = √ = √ = 2. 00 × 1 0 − 1 3 m
2 d F 2 d π F/ 2 d π F ( 3. 1 8 × 1 0 − 4 m) π 3. 996 × 1 0 6
8-1 0. ( a) The mode number is,

2 n d ( 2) ( 4. 5) ( 2 cm)
m= = = 330, 000
λ 5. 46 × 1 0 − 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 Tm in Tm in ( 1 − r2 ) 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 × 1 0 6
λ
8-1 1 . The spacing d at which overlap occurs is, d = m λ 2 / 2 = ( m + 1 ) λ 1 / 2. Thus,

m λ2 − λ1 ) = λ1
2d
∆λ = λ
λ
λ 2
( 490 × 1 0 )
−9 2
d= = m = 0. 021 8 m = 2. 1 8 cm
2 ∆λ 2 ( 0. 0055 × 1 0 − 9 )
8-1 2 . The bright bands due to white light satisfy, m λ = 2 d. Thus,

m1 λ 1 = m2 λ 2 = 2 d and m 1 − m 2 = 1 49
These equations solved simultaneously gives m 2 = 588. 7 → 588. Then,
d = m 2 λ 2 / 2 = ( 588) ( 546. 1 × 1 0 − 9 nm) / 2 = 0. 1 61 mm
44
8-1 3. The irradiance as a function of phase has the form I = 4I0 cos 2 ( δ/ 2) .
( a) The fringe maximum occur for,
cos δm / 2 = ± 1 , δm / 2 = m π ⇒ δ m = 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 ± δF W / 2
cos 2 ( δ/ 2) = 1 / 2
cos 2 ( ( δm ± δF W ) / 2) = cos 2 ( ( 2 m π ± δF W / 2) / 2) = cos 2 ( δF W / 4) = 1 / 2
√
cos( δF W / 4) = ± 1 / 2
| δF W / 4| = π/ 4 ⇒ δF W = π
δm + 1 − δm 2 π
( c) The finesse is then, F = = = 2.
δF W π
8-1 4. Case 1 : The irradiances of the emerging beams are
I1 = ( 0. 2) ( 0. 8) I0 = 0. 1 6 I0 and I2 = ( 0. 8) ( 0. 2) I0 = 0. 1 6 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 = 1 6 I1
The visibility is,

p
2 1 6 I12 8
V= = = 0. 47
I1 + 1 6 I1 1 7
8-1 5. ( a) Following the method of Section 8 5 and using Eq. ( 8 22) :

r t2 e − i δ − r + r ( r2 + t2 ) e − i δ
E0 R = − r E0 I + r t e − i δ E0+1 = − r+ −
E0 I = E0 I
1−r e 2 i δ 1 − r2 e − i δ
r( 1 − e − i δ ) 2 i r e − i δ / 2 sin δ/ 2
E0 R = E 0 I =
1 − r2 e − i δ 1 − r2 e − i δ
E0 R E0∗ R 4r 2 sin 2 δ/ 2
R= =
E0 I E0∗ I ( 1 − r 2 ) 2 + 4r 2 sin 2 δ/ 2
Note that I have assumed lossless mirrors in setting r 2 + t 2 = 1 .
4r 2 sin 2 δ/ 2 + ( 1 − r 2 ) 2
( b) R + T = =1
( 1 − r 2 ) 2 + 4r 2 sin 2 δ/ 2
8-1 6. Using the relation between the finesse and the ratio of the maximum and minimum transmission,
Tm a x − Tm in Tm ax T 4 r2 4 ( 0. 6)
F= = − 1 ⇒ m ax = 1 + F = 1 + =1+ = 16
Tm in Tm in Tm in ( 1 − r2 ) 2 ( 1 − 0. 6) 2
45
√
8-1 7. In this case, in steady state, Eq. ( 8 21 ) should be written as, E0+1 = t 1 E0 I + E0+1 R 1 R 2 e − i δ
p
where δ = 2 kd + ϕ 1 R + ϕ 2 R and r 1 , 2 = R 1 , 2 e − i ϕ 1 , 2 R . Rearranging,
t1
E0+1 = √ E0 I
1− R1 R2 e − i δ
The complex amplitude of the transmitted field is then obtained in a manner that mirrors the development
leading to Eq. ( 8 23) ,
t1 t2 e − i δ / 2
E0 T = √ E0 I
1 − R1 R2 e − i δ
Then,
I E E∗ | t1 | 2 | t2 | 2 T1 T2
T = T = 0 T 0∗ T = √ = √ √
II E0 I E0 I 1 + R 1 R 2 − 2 R 1 R 2 cos δ ( 1 − R 1 R 2 ) 2 + 4 R 1 R 2 sin 2 ( δ/ 2)

Here T1 , 2 = t 1 , 2 | 2 . If the mirrors are lossless, then T1 , 2 = 1 − R 1 , 2 and R = 1 − T.
FSR
8-1 8. ( a) Estimating from Figure 8 1 5, F = ≈ 70 ( b) ν fs r = c/ 2d = ( 3 × 1 0 8 ) / ( 2 · 0. 1 ) Hz = 1 . 5 GHz
FWHM
νfs r 1 . 5 GHz ν 4. 53 × 1 0 1 4
( c) FWHM = = = 21 MHz ( d) Q = = = 2. 2 × 1 0 7
F 70 FWHM 2. 1 × 1 0 7
( e) τp ≈ Q / ( 2 π ν) = ( 2. 2 × 1 0 7 ) / 2 π · 4. 53 × 1 0 1 4 ) ∼ 8 × 1 0 − 9 s
8-1 9.
The transmittance function has the form, 1
0.9
1 0.8
T= 4 F2
1+ π2
sin 2 ( k d) 0.7
0.6
For the case at hand, 0.5
T
0.4
0.3
F = 1 5 and k = 2 π/ λ = 2 π/ ( 632. 8 × 1 0 − 9 m) = 9. 93 × 1 0 6 / m.
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)
8-2 1 . ( a) The peak separation for a given wavelength is d fs r = λ/ 2 ⇒ λ = 2 ∆ d = 2 · 2. 95 µm = 5. 9 µm
∆d 0. 01 × 1 0 − 6
( b) Using Eq. ( 8 37) , ∆λ ≈ λ= 5. 9 µm = 1 . 2 × 1 0 − 6 µm
d 0. 05
2 2 2 m 1 d 2 − 2 m 2 d 1 2 m 1 ( d 1 + ∆ d) − 2 ( m 1 + 1 ) d 1 2 m 1 ∆d − 2 d 1
( c) λ 2 − λ 1 = d2 − d1 = = ≈
m2 m1 m1 m2 m1 ( m1 + 1 ) m 21
Now, m 1 = 2 d 1 / λ ≈ 2 d/ λ. So,

∆d λ2 5. 9 2
λ2 − λ1 = λ − = 1 . 2 × 1 0 − 6 µm − µm = − 3. 5 × 1 0 − 4 µm.
d 2d 2 · 5 × 1 0− 4
8-2 2 . ( a) From the plot, 28 divisions ≈ 1 1 . 6 GHz. The separation of the absorption dips is then,
1 1 . 6 GHz
∆ νd ip s ≈ ( 25. 5 divisions) = 1 0. 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
8-2 3. 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 + δλ 1 / 2
1 1 λ + δλ 1 / 2 − λ 1
− = =
λ λ + δλ 1 / 2 λ ( λ + δλ 1 / 2 ) 4dF
δλ 1 / 2 λ λ λ
≈ δλ 1 / 2 = =
1 + δλ 1 / 2 / λ 4dF 2mF
2 δλ 1 / 2 ≈ λ/ ( mF)
47
C hapter 9 C oherence
9-1 . The given

function is
− 1, − L/ 2 < x < 0
f( x) = Odd function: a 0 = a n = 0
+ 1, 0 < x < L/ 2
Its Fourier series representation is formed as,
∞
X Z
2 + L/2
f( x) = b m sin mkx , where b m = f( x) sin mkx dx =
L − L/2
m= 1
( Z ZL/2 )
2 0
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 2 πmx 2 2 πx 2 6πx
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,



 3 5
 E cos ω t , π , ω t, π
0
f( t) = 2 2

 5 7
 0, < ωt< π
2π 2
The Fourier series representation is formed as:
∞ Z Z t0 + T
a0 X 2 t0 + T 2
f( t) = + a m cos mωt where a 0 = f( t) dt and a m = f( t) cos mωt dt
2 T t0 T t0
m= 1
Z t0 + T Z # 5 π/ 2
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
3 π/ 2
Z 5 π/ 2
2 E0
am = cos ωt cos mωt d( ω t)
ωT 3 π/ 2
Z 5 π/ 2 5 π/ 2
2 E0 2 E0 ω t 2 ω t E
m = 1 : a1 = cos 2 ωt d( ω t) = + = 0
ωT 3 π/ 2 2π 2 4 3 π/ 2 2
5 π/ 2 5 π/ 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: a 2 = + =
ωT 2( m − 1 ) 2( m + 1 ) 3 π/ 2 2 π 2 6 3 π/ 2 3π
5 π/ 2
E0 sin 3 ωt sin 5ωt 2 E0
m = 3 : a 3 = a 5 = a 7 = . . . = 0, m = 4: a 4 = + =−
π 6 10 3 π/ 2 15 π
All b m = 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,

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

2
t2 t iσω σ 2 ω2
− i ωt= √ − √ +
2 σ2 2σ 2 2
Then,
" 2 #
2
Z∞ − √
t
−
iσω
√ +
σ2 ω2 Z∞ t iσω
h 2σ 2 2 h σ 22ω 2 − √
2σ
− p
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
Z Z∞
h − σ22 ω 2 ∞ − x 2 √ h − σ22 ω 2 √ 2 h − σ22 ω 2 √ √
g( ω) = e e σ dx = e 2 σ e − x dx = e 2 σ π
2π −∞ 2 π −∞ 2 π
σh − σ 2 ω 2 / 2
g( ω) = √ e
2π
9-4. In general,
Z∞ Z∞
1
f( t) = g( ω) e iωt
dω where g( ω) = f( t) e i ω t dt
−∞ 2π −∞
For the given function,

Z τ0 / 2 i ω t τ0 / 2 " #
1 A e A e i ωτ0 / 2
− e − i ωτ0 / 2
A ωτ Aτ
0 0
g( ω) = A e i ω t dt = = = sin = sinc ( τ0 / 2)
2π − τ0 / 2 2 π i ω − τ0 / 2 π ω 2i πω 2 2π
2 A 2 τ02
| g( ω) | = sinc 2 ( 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, l t = λ 2 / ∆λ. Then, ( l t ) 1 = = 3. 48 × 1 0 − 4 m and ( l t ) 2 = = 3. 48 × 1 0 − 5 m.
1 00 10
Use longer coherence − length light, thus shorter − bandwidth filter.
9-6. The coherence length is l t = cτ0 = c( 1 0 − 6 s) = 300 m. The bandwidth is, ∆ ν = 1 / τ0 = 1 0 6 Hz. The resultant
linewidth is ∆λ = λ 2 / l t = ( 632. 8 × 1 0 − 9 m) 2 / ( 300 m) = 1 . 33 × 1 0 − 1 5 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 × 1 0 − 7 cm)
ls < = = 0. 00769 cm
θ 0. 008727
If l s is the diameter of a circle of coherence, then “good” coherence exists over an area of 1 0% the area, or a
√ √
diameter of 0. 1 l s < . 1 ( 0. 0769 mm) = 0. 0243 mm.
λ 2 ( 643. 8 nm) 2 l 30 cm
9-8. ( a) ∆λ = = = 0. 001 38 nm ( b) τc = t = = 1 ns
lt 30 × 1 0 7 nm c c
λ2 ˚ )2
( 5000 A
9-9. The coherence length of the transmitted light is l t = = = 0. 25 cm = 2. 5 mm.
∆λ ˚
1A
˚ / mm) ( 0. 2 mm) = 4 ˚A . The coherence length and time

9-1 0. The linewidth is for the 0. 2 mm exit slit is ∆λ = ( 20 A
are then,
λ2 ( 5 × 1 0 − 5 cm) 2
lt = = = 0. 0625 cm
∆λ 4 × 1 0 − 8 cm
τ0 = l t / c = 2. 08 × 1 0 − 1 2 s
Lens 2
Lens 1
Prism Slit 2
Slit 1
9-1 1 . See Figure 9 1 4 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 × 1 0 − 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( 1 00 cm) ( 5. 89 × 1 0 − 5 cm)
a < ls = = = = 0. 1 44 cm
θ s 0. 05 cm
λ 2 ( 6328 × 1 0 − 1 0 m) 2 1 c c
9-1 2 . ( a) ∆λ = = = 4 × 1 0 − 7 ˚A , ∆ν = = = 4 = 3 × 1 0 4 Hz
lt 1 04 m τ0 l t 1 0 m
˚ / mm) ( 0. 2 mm) = 4 A
9-1 3. ( a) The linewidth is for the 0. 2 mm exit slit is ∆λ = ( 20 A ˚ . The coherence length and
time are then ( as in problem 9 1 0) ,
λ2 ( 5 × 1 0 − 5 cm) 2
lt = = = 0. 0625 cm
∆λ 4 × 1 0 − 8 cm
τ0 = l t / c = 2. 08 × 1 0 − 1 2 s
0. 4 mm
( b) The normalized correlation function is | γ( τ) | = 1 − τ/ τ0 . But, τ/ τ0 = ∆l / l t = = 0. 64. So,
0. 625 mm
V = | γ( τ) | = 1 − τ/ τ0 = 0. 36.
Im a x − Im in 1 00 − Im in
( c) V = = 0. 36 = ⇒ Im in = 47. 06.
Im a x + Im in 1 00 + Im in
50
9-1 4. The coherence length l t and spatial coherence width l s are the length and diameter of the cylindrical coher
ence volume. Now,
λ2 550 2
lt = = nm = 1 . 0083 × 1 0 − 4 cm = 1 . 83 λ ave
∆ λ 700 − 400
1 . 22λ 1 . 22( 550)
ls = = nm = 76890 nm = 7. 689 × 1 0 − 3 cm
θ 0. 008727
“Good” l s = 0. 25( 7. 689 × 1 0 − 3 ) = 1 . 922 × 1 0 − 3 cm = 35 λ ave
The area of the base of the cylindrical coherence volume is
π l s2 π
= ( 1 . 922 × 1 0 − 3 ) 2 = 2. 90 × 1 0 − 6 cm 2
A b a se =
4 4
√
9-1 5. ( a) For unequal beams and partial coherence, I = I1 + I2 + 2 I1 I2 Re( γ) where,

Re( γ) = Re ( 1 − τ/ τ0 ) e i ω τ = ( 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 − Im in I1 + I2 + 2 I1 I2 | γ( τ) | − I1 + I2 − 2 I1 I2 | γ( τ) | ) 2 I1 I2 | γ( τ) |
V= = √ √ =
Im ax + Im in I1 + I2 + 2 I1 I2 | γ( τ) | + I1 + I2 − 2 I1 I2 | γ( τ) | ) I1 + I2
( b) For beams of the same irradiance √
2 I1 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 R 2 − 2. 38 R + 0. 81 = 0 ⇒ R = 2. 55
τ ∆ ∆ mλ m ∆λ
9-1 6. The optical path difference in m t h order is ∆ = m λ. Then, = = 2 = =
τ0 l t λ / ∆λ λ 2 / ∆ λ λ
m ∆λ
As shown in problem ( 9 1 5b) for equal irradiance beams, V = | γ( τ) | = 1 − τ/ τ0 = 1 −
λ
9-1 7. ( a) l t = λ 2 / ∆λ = ( 546. 1 ) 2 / 0. 05 = 5. 965 × 1 0 6 nm

m ∆λ 20 ( 0. 05)
( b) With a filtered Hg lamp, V = 1 − =1− = 0. 998.
λ 546. 1
20 ( 1 0)
With filtered white light, λ ave = 546 nm and ∆ λ = 1 0 nm, V = 1 − = 0. 63
546
∆λ ( ∆ ) ( ∆ λ) ( 2 × 1 0 7 ) ( . 001 3)
9-1 8. ∆ = m λ = 2 × 1 cm = 2 × 1 0 7 nm, V = 1 − m =1− =1− = 0. 937
λ λ 2 ( 643. 847) 2
( 1 0 8 ) ( 0. 001 3)
For a 5 cm movement, ∆ = 2 × 5 cm = 1 0 8 nm, and V = 1 − = 0. 686
( 643. 847) 2
( 2 d) ( ∆λ)
V → 0 when 1 − → 0 or when d = λ 2 / 2 ∆λ, so d = 0. 0895 cm.
λ2
( 2 × 1 0 7 ) ( 0. 025)
9-1 9. ( a) d = 1 cm: V = 1 − < 0 or V = 0 ( See Prob. 9 1 6) . d = 5 cm, V = 0
( 546. 1 ) 2
λ2 ( 546. 1 ) 2
V = 0 first when d = = = 0. 596 cm
2 ∆λ 2 ( 0. 025)

∆λ λ2 ( 0. 1 5) ( 546. 1 ) 2
( b) V = 0. 85 = 1 − 2 d 2 ⇒ d = 0. 1 5 = = 0. 0895 cm
λ 2 ∆λ 2 ( 0. 025)
51
C hapter 1 0 Fiber O ptics
1 0-1 . The information rate is,

( 4000 Hz) ( 2 × sampling rate) ( 8 bits/ sample) = 64kbps
44. 7 Mbps
Number of channels = − 26( assigned) = 672
64 kbps
1 0-2 . Each TV channel requires a bandwidth of 6 MHz.
ν c/ λ ( 3 × 1 0 8 / 1 . 55 × 1 0 − 6 )
( a) # of TV stations = = = = 32 million
6 × 1 0 Hz 6 × 1 0 Hz
6 6 6 × 1 06
4 × 1 01 2
( b) # of TV stations = = 0. 67 million
6 × 1 0 6 Hz
r
n1 2 p
1 0-3. ( a) In general, L s = d − 1 . When θ = θ m , sin θ = sin θ m = n 21 − n 22 . Then,
sin θ
s 2
n1 n d
Ls = −1=p 2
n 21 − n 22 n 21 − n 22
( b) In one meter of fiber, there are 1 / L s reflections, so the number of reflections per meter Ns is,
p
n 21 − n 22 ( 1 . 46 2 − 1 . 457 2 ) 1 / 2
Ns = = = 1 284/ m
n2 d 1 . 457 ( 50 × 1 0 − 6 m)
1 0-4. ( a) 1 . 52 sin ϕ c = 1 . 41 ⇒ ϕ c = 68. 1 ◦ ( b) N. A. = n 0 sin θ m = n 1 cos ϕ c = 1 . 52 cos( 68. 1 ◦ ) = 0. 567

( c) sin θ m = ( N. A. ) / n 0 = ( 0. 567/ 1 ) = 0. 567 ⇒ θ m = 34. 5 ◦
p √
1 0-5. ( a) N. A. = n 21 − n 22 = 1 . 53 2 − 1 . 39 2 = 0. 64
( b) Since n 0 sin θ m = N. A. , θ m = sin − 1 ( 0. 6394) = 39. 7 ◦ . The cone angle is 2 θ m = 79. 5 ◦ .
( c) The number of reflections per inch is Ns = 1 / L s , so,
1 1 1
Ns = = r 2 = r 2 = 1 84/ inch
Ls n1 1 . 53
d n 0 sin θ
−1 0. 0025 in) sin ( 3 9 . 7 5 ◦ )
−1
That is, there will be 6624 reflections in 3 ft. At θ = θ m / 2 = 1 9. 87 ◦ , 1 / L s = 91 . 1 4/ inch so that there would
be 3281 reflections in 3 ft.
1 0-6. Refer to Figure 1 0 3b in the body of the text. In that figure, x s is the hypotenuse of the triangle formed by
θ 0 , L s , 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 0 n 0 sin θ sin θ
( b) The total distance is x t = x s Ns where Ns is the total number of skips. That is,
" #
L n1 d L n1 L
xt = xs = p = 1 /2
L s sin θ d ( n 1 / sin θ) 2 − 1 n 1 − sin 2 θ
2
( c) For the given parameters,

n 1 d ( 1 . 5) ( 50 µm)
xs = = = 432 µm
sin θ sin( 1 0 ◦ )
q p
Ls = d n 1 / sin θ) 2 − 1 = ( 50 µm) ( 1 . 5/ sin( 1 0 ◦ ) ) 2 − 1 = 429 µm
n1 L 1 . 5 ( 1 0 m)
xt = 1 /2 = p = 1 0. 07 m
n 21 − sin 2 θ 1 . 5 2 − sin 2 ( 1 0 ◦ )
52
1 0-7. The number of modes is,
2 2
1 πd 2 1 πd 1
m m ax = ( ( N. A. ) = n 21 − n 22 = ( π 40/ 0. 85) 2 1 . 461 2 − 1 . 456 2 = 1 59
2 λ 2 λ 2
1 0-8. The necessary condition is,

d 2. 4 2. 4 λ 2. 4 ( 1 . 25 µm)
< ⇒d= = √ = 1 0. 2 µm
λ π ( N. A. ) π ( N. A. ) π 1 . 46 2 − 1 . 457 2
p
1 0-9. In general the number of √ modes is given by N = 1 + ( d/ λ) n 21 − n 22 . Then,
For d = 5 λ: N = 1 + 2 ( 5) 3. 6 2 − 3. 55 2 = 6 ( or 1 2 for two polarizations) .
√
For d = 50λ: N = 1 + 2 ( 5) 3. 6 2 − 3. 55 2 = 60 ( or 1 20 for two polarizations) .
1 0-1 0. The loss in db/( 1 00 m) is

α d b = 1 0 db/ ( 1 00 m) log 1 0 ( P1 / P2 ) = ( 0. 1 db/ m) log 1 0 ( 5 ) = 0. 070 db/ m = 70 db/ km
1 0-1 1 . The entire cable has a loss of 3 × 5 db + 2 db = 1 7 db in 3 km. So α d b = 1 7db/ 3 km. Then, using Eq. ( 1 0 1 3) :

α d b = ( 1 0 db/ km) log 1 0 ( P1 / P2 ) ⇒ P2 = P1 1 0 − α d b L / 1 0 ( d b / k m )

P2 = ( 4 mW 1 0 − [ ( 1 7 / 3 ) 3 / 1 0 ] = ( 4 mW 1 0 − 1 7 / 1 0 = 0. 080 mW
1 0-1 2 . The loss in db that can be tolerated in a length L of fiber is,

α d b L = ( 1 0 db) log 1 0 ( 0. 01 / 1 0 − 6 ) = 40 db
So for the coaxial cable: L = ( 40 db/ 1 2 db/ km) = 3. 33 km. For the fiber: L = ( 40 db/ 4 db/ km) = 1 0 km.
1 0-1 3. Rayleigh scattering has a 1 / λ 4 dependence so,

α 1 . 5 5 = α 0 . 9 0 0. 9/ 1 . 55) 4 = ( 1 . 2 db/ km) 0. 9/ 1 . 55) 4 = 0. 1 36 db/ km
1 0-1 4. ( a) The fractional power loss is

P1 − P2
f= = 1 − P2 / P1 ⇒ P2 / P1 = 1 − f
P1
Then,
1
α d b = ( 1 0 db/ km) log 1 0 = − ( 1 0 db/ km) log1 0 ( 1 − f)
1−f
( b) Using the relation from part ( a) , For f = 0. 25, α d b = 1 . 25 db/ km, for f = 0. 75, α d b = 6. 02 db/ km ,
forf = 0. 90, α d b = 1 0db/ km and for f = 0. 99, α d b = 20 db/ km
1 0-1 5. ( a) The ratio of the longest and shortest path lengths is,
L sh ort n 2 1 . 45
= =
L lon g n 1 1 . 46
so,
L lon g = L sh ort ( 1 . 46/ 1 . 45) = ( 1 km) ( 1 . 46/ 1 . 45) = 1 . 0069 km
( b) The transit times are ,
L s h ort n 1 1 . 46
t s h ort = = L s h ort = ( 1 km) = 4. 867 × 1 0 − 6 s
v c 3 × 1 0 5 km/ s
L n 1 . 46
t s h ort = s h ort = 1 L s h ort = ( 1 . 0069 km) = 4. 900 × 1 0 − 6 s
v c 3 × 1 0 5 km/ s
53
1 0-1 6. First find the index of refraction of the cladding:
p p p
sin θ m = n 21 − n 22 ⇒ n 2 = n 21 − sin 2 θ m = 1 . 446 2 − sin 2 ( 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 × 1 0 − 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
1 0-1 7. The entrance angle θ = 1 5 ◦ is not necessarily θ m so Eq. ( 1 0 1 4) ,

! " #
∆L n 1 n1 n1 n1 ( 1 . 48) ( 1 km) 1 . 48
δ( τ) = = ∆L = ( x t − L) = L p −1 = p −1
v c c c n 21 − sin 2 θ 3 × 1 0 − 4 km/ ns 1 . 48 2 − sin 2 ( 1 5 ◦ )
δ( τ) = 77. 2 ns
Here the expression for x t is given by problem 1 0 6b.
1 0-1 8. The group delay is

n n1 − n2 1 . 46
δ( τ/ L) = 1 = ( 0. 003) = 1 4. 6 ns/ km
c n2 3 × 1 0 − 4 km/ ns
1 0-1 9. Using Eq. ( 1 0 1 5) the refractive index profiles is seen to be,

p p
n( r) = n 1 1 − 2 ∆( r/ a) α = 1 . 5 1 − ( 0. 02) ( r/ 50 µm) α
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)
1 0-2 0. The relative index difference is,

n 1 − n 2 1 . 46 − 1 . 44
∆= = = 0. 01 37
n2 1 . 46
Using Eq. ( 1 0 1 6) and the relation that follows,
n1 L 2 1 . 46 ( 1 km)
δ( τ) G = ∆ = ( 0. 01 37) 2 = 457 ps
2c 2 ( 3 × 1 0 − 4 km/ ns)
2 2
δ( τ) SI = δ( τ) G = δ( τ) G = 1 46 × 457 ps = 66. 7 ns
∆ 0. 01 37
0. 5 0. 5 0. 5
1 0-2 1 . νm a x L = ⇒ νm ax = = = 25 MHz
δ( τ/ L) L δ( τ/ L) ( 1 km) ( 20 ns/ km)
1 0-2 2 . At λ = 820 nm, Figure 1 0 1 3 gives M ≈ 1 00 ps/ nm km. Then δ( τ/ L) = − M ∆ λ. So,
For the LED: δ( τ) = − M ∆λ L = ( 1 00) ( 40) ( 1 ) ps = 4000 ps = 4 ns.

For the LD: δ( τ) = − M ∆λ L = ( 1 00) ( 4) ( 1 ) ps = 400 ps = 0. 4 ns.
54
1 0-2 3. 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 × 1 0 − 4 km/ ns
δτm at = M L ∆λ = ( 1 00) ( 1 ) ( 40) ps = 4 ns
So,
√
δτ = 48. 7 2 + 4 2 ns = 48. 9 ns
1 0-2 4. ( a) The plot is shown below,
8
Wavegu de d spers on (ps/km/ns)
7 ( b) From the plot:

M 0 ≈ 3. 9 ps/ km nm at 1 . 27 µm
6
M 0 ≈ 4. 3 ps/ km nm at 1 . 55 µm
5 Then,
At 1 . 27 µm:
4 ps ps
∆ ( τ/ L) = M 0 ∆λ = ( 3. 9) ( 1 ) = 3. 9
km km
3 At 1 . 55 µm:
ps ps
∆ ( τ/ L) = M 0 ∆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
1 0-2 5 . ( a) Modal:
n1 L n1 − n2 ( 1 . 47) ( 1 ) 1 . 47 − 1 . 455
δτm o d = = ns = 50. 5 ns
c n2 3 × 1 0− 4 1 . 455
Material:
δτm at = L M ∆ λ = ( 1 ) ( 43) ( 25) ps = 1 . 075 ns
Waveguide:
δτW G = L M 0 ∆λ = ( 1 ) ( 3 ) ( 25) ps = 0. 075 ns
( b) The total is then,
p
δτ = δτm2 o d + δτm2 at + δτW
2
G ≈ 50. 5 ns
1 0-2 6. Consider Figure 1 0 1 6a 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
∆ ϕ 1 = k ∆ L and that for the light√ exiting through√ Output 2 is ∆ ϕ 2 = k ∆ L + π, where the addition of π is
due to the fact that r 10 = − 1 / 2 = e i π 1 / 2 = e i π r 2 . Now the transmission coefficients at each beam
√
splitter must be t 1 = t 2 = 1 / 2 so that the amplitudes each of the four individual beams exiting through an
output port is
p p 1
E0 , ou t = ( 1 / 2) ( 1 / 2) E0 , in = Ein
2
The irradiance of each of the four individual output beams is then, Iou t = Iin / 4. The total irradiance at
each output port is then given by the two beam, equal irradiance, interference relation,
I I
IT , p ort 1 = 2 Iou t ( 1 + cos( ∆ ϕ 1 ) ) = 2 in [ 1 + cos( k ∆L) ] = in [ 1 + cos( k ∆ L) ]
4 2
Iin I
IT , p ort 2 = 2 Iou t ( 1 + cos( ∆ ϕ 2 ) ) = 2 [ 1 + cos( k ∆ L + π) ] = in [ 1 − cos( k ∆L) ]
4 2
Thus the total irradiance through the two outputs is, IT = IT , p ort 1 + IT , p ort 2 = Iin .
55
∆λ 0. 8 3 × 1 08
1 0-2 7. ( a) The frequency difference is ∆ν = ν= Hz = 1 0 0. 0 GHz
λ 1 550 1 550 × 1 0 − 9
( b) 40 × 1 0 GHz = 4 THz.
1 0-2 8. ( a) The conditions to be satisfied are,
∆L = m λ 1 n = ( m + 1 / 2) λ 2 n
Here λ 1 , 2 n = λ 1 , 2 / n = λ 1 , 2 / 1 . 5. No ∆L satisfies this relation with exactly integer m but a reasonably good
solution is
∆ L = 950. 01 λ 1 n = ( 950. 01 ) ( 1 550. 8 nm) / 1 . 5 = 950. 500 λ 2 n
Many other approximate solutions exist.

( b) Since,

∆L 950. 01 λ 1 n 1 550. 8
= = 950. 01 = 949. 5201 778808971
λ3n λ 3n 1 551 . 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 1 0 26:
IT , 2 1 − cos( 2 π∆ L/ λ n )
R= =
IT , 1 1 + cos( 2 π ∆ L/ λ n )
Here,
2 π ∆L/ λ n = 2 π( 950. 01 ) λ 1 n / λ 2 n = ( 2 π) ( 950. 01 ) ( 1 550. 8/ 1 550. 4) = 1 900. 51 π
and
R = 1 . 065 ⇒ 1 / R = 0. 94
56
C hapter 1 1 Fraunhofer Diffraction
1 1 -1 . See Figure 1 1 1 8 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 θ m = b ym / f ⇒ ym = m λ f/ b
( a) The first minimum occurs at y1 = λ f/ b = ( 546. 1 × 1 0 − 6 mm) ( 60 cm) / ( 0. 01 5 cm) = 2. 1 8 mm.

( b) The separation of the first and second minimum is
y2 − y1 = ( 2 − 1 ) λ f/ b = 2. 1 8 mm
1 1 -2 . Refer to Figure 1 1 1 and the surrounding discussion. The ratio I/ I0 is given be Eq. 1 1 1 0 is
sin 2 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 sin 2 ( 3 π/ 4)
= = 0. 090
I0 ( 3 π/ 4) 2
1 1 -3. See Figure 1 1 1 9 that accompanies the problem in the text.

( a) The diffraction minima are located at angles θ m = ym / L where L = 2 m is the slit to screen distance,
The positions of the minima are given by m λ = b sin θ m = bym / L ⇒ ym = mλL/ b. Then,
6 λ L 6 ( 632. 8 × 1 0 − 7 cm) ( 200 cm)

y3 − y − 3 = ∆ y = ( 3 − ( − 3 ) ) λ L/ b ⇒ b = = = 0. 01 3 cm = 0. 1 3 mm
∆y 5. 625 cm
( b) L m in = b 2 / 2λ, so,
L 200 cm
= = 1 39
L m in ( 0. 01 35 cm) 2 / ( 2 · 632. 8 × 1 0 − 7 cm)
The screen is in the far field.
1 1 -4. Let m 1 = 5 for λ 1 and m 2 = 4 for λ 2 . Then,
m 1 λ 1 = m 2 λ 2 = b sin θ
5 λ 1 = 4 λ 2 = 4 ( 620 nm) ⇒ λ 1 = 496 nm
1 1 -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. 1 25 µm, for ϕ = 45 ◦ , b = 1 . 437 µm, for ϕ = 90 ◦ , b = 0. 778 µm, for ϕ = 1 80 ◦ , b = 0. 55 µm.
57
( 1 ) ( 550)
1 1 -6. ( a) sin θ = m λ/ b = ⇒ θ = 1 5◦
21 25
kb πb π ( 2. 1 25 µm)
( b) β = sin θ = sin θ = sin θ = 1 2. 1 38 sin θ. Then
2 λ 0. 55 µm
I/ I0 = sin 2 β/ β 2
θ=5 : ◦
β = 1 . 058 rad, I/ I0 = 0. 678
θ = 1 0◦ : β = 2. 1 08 rad, I/ I0 = 0. 1 66
θ = 1 5◦ : β = 3. 1 42 rad, I/ I0 ≈ 0
θ = 22. 5 ◦ : β = 4. 645 rad, I/ I0 ≈ 0. 0461
1 1 -7. ( a) The fourth and fifth intersections of β = tanβ occur at 4. 477 π and 5. 482 π.
Iβ = 4 . 4 7 7 π sin 2 ( 4. 477 π) / ( 4. 477 π) 2 1 Iβ = 5 . 4 8 2 π sin 2 ( 5. 482 π) / ( 5. 482π) 2 1

( b) = = ; = =
Iβ = 0 1 1 99 Iβ = 0 1 298
1 1 -8. For the single slit:

I1 sin 2 ( 1 . 43 π) / ( 1 . 43π) 2
= = 0. 047
I0 1
I2 sin 2 ( 2. 46 π) / ( 2. 46π) 2
= = 0. 01 7
I0 1
For the circular aperture: ( See Figure 1 1 8b)
I1
= 0. 01 75
I0
I2
= 0. 0041 6
I0

k πD πD y fλ γ ( 1 706. 9) ( 5. 5 × 1 0 − 5 )
1 1 -9. γ = D sin θ = sin θ = ⇒ y= =γ cm = ( 3. 268 × 1 0 − 4 cm) γ
2 λ λ f πD 91 . 44π
1 st secondary maximum ( γ = 5. 1 5) : y = 1 . 68 × 1 0 − 3 cm
2 n d secondary maximum ( γ = 8. 42) : y = 2. 75 1 0 − 3 cm
1 1 -1 0. 1 . 22 D sin θ = D y/ f ⇒ y = R = 1 . 22 λ f/ D = ( 1 . 22) ( 5. 5 × 1 0 − 5 cm) ( 1 50) / 1 2 = 8. 39 × 1 0 − 4 cm
1 1 -1 1 . Using Eq. ( 1 1 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 ( 1 0. 6 × 1 0 − 6 m) ( 3. 76 × 1 0 8 m)
R = L tan ∆ θ 1 / 2 ≈ L ∆θ 1 / 2 = = = 4. 86 × 1 0 6 m
D 1 0− 3 m
The diameter of the laser spot on the moon is about 9. 72 × 1 0 6 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 × 1 0 − 1 1 W/ m 2
A π R 2 π ( 4. 86 × 1 0 6 m) 2
58
1 1 -1 2 .
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 × 1 0 − 3 m) 2
θ≈ = 1 . 22 λ/ D ⇒ L = = = 5. 1 8 m
L 1 . 22 λ 1 . 22 ( 632. 8 × 1 0 − 9 m)
L
1 1 -1 3. The distance L for the headlights to be barely resolvable if they are separated by a distance y is given be
Eq. ( 1 1 22) , as,
yD ( 45 × 2. 54 cm) ( 0. 5 cm)
∆ θ m in = y/ L = 1 . 22 λ/ D ⇒ L = = = 8. 51 7 × 1 0 5 cm = 27, 900 ft = 5. 3 miles
1 . 22 λ 1 . 22 ( 5. 5 × 1 0 − 5 cm)
1 1 -1 4. ( a) The minimum and maximum distances are for a line separation of s = 1 mm is

s 1 . 22 λ s D m in ( 1 0 − 3 ) 2 × 1 0 − 3
Min: ∆θ = = ⇒L= = m = 3. 0 m
L D 1 . 22 λ 1 . 22 · 550
× 1 0− 9
s Dm ax ( 1 0 − 3) 7 × 1 0 − 3
Max: L= = m = 1 0. 4 m
1 . 22 λ 1 . 22 · 550 × 1 0 − 9
( b) The pupil diameter will be
1 . 22 λ L 1 . 22( 550 × 1 0 − 9 ) L
D= = = 6. 71 × 1 0 − 4 L
s 1 0− 3
1 1 -1 5. ( a) According to Eq. ( 1 1 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 cos 2 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 cos 2 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
πb λ I sin β sin ( π/ 4)
p = 1 : sin θ = λ/ a; β = = π ( b/ a) ; = = = 0. 81 06
λ a 4 I0 β π/ 4
2 2
π b 2λ I sin β sin ( π/ 2)
p = 2: sin θ = 2λ/ a; β = = 2π ( b/ a) ; = = = 0. 4053
λ a 4 I0 β π/ 2
2 2
π b 3λ I sin β sin ( 3 π/ 4)
p = 3 : sin θ = 3 λ/ a; β = = 3 π ( b/ a) ; = = = 0. 0. 0901
λ a 4 I0 β 3 π/ 4
1 1 -1 6. ( a) If a/ b = N where N is an integer, the the N t h fringe is missing on either side of the central maximum.
If a/ b falls between N − 1 and N, then the ( N − 1 ) t h 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 1 3 bright fringes in the central diffraction peak,

2 ( a/ b) − 1 = 1 3 ⇒ a/ b = 7 ⇒ a = 7 b = 7 ( 0. 3 mm) = 2. 1 mm
59
1 1 -1 7. ( 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 ( ∆θ 1 / 2 ) d iff = λ/ b.
Interference minima occur for,
( p + 1 / 2) λ = a sin θ ≈ a θ ⇒ ( ∆θ 1 / 2 ) int = λ/ 2 a ( p = 0)
Thus,
( ∆θ 1 / 2 ) d iff λ/ b
= = 2 ( a/ b)
( ∆θ 1 / 2 ) int λ/ 2a
( b) For a = 1 0 b:
peak width
= 2 ( a/ b) = 20 ×
fringe width
1 1 -1 8.
Consider the geometry shown at left. Begin with an expression like Eq. ( 1 1 4)
s but for three slits. Then,
( Zb Z − a+ b Za+ b )
EL i ( k r 0 − ω t) 2
i k s sin θ
2
i k s sin θ
2
i k s sin 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 Zb
2 k
a e i k s s in θ ds = b sin( β) / β as for the single slit in the text. Here β = b sin θ.
−b 2
2
Z − a+ b
2 1 h i k s in θ ( a + b/ 2 ) i e2iα
e i k s s in θ ds = e − e i k sin θ ( a − b / 2 ) = ( 2 sin β)
− a−
b i k sin θ k sin θ
2
k
where α = a sin θ. The final integral follows from the one just done by letting a → − a , α → − α. That is
2
Za+ b
2 e − 2iα
e i k s sin θ ds = ( 2 i sin β)
a− 2
b i k sin θ
Altogether, then,

EL i ( k r 0 − ω t) e2iα e − 2iα
EP = e b sin( β) / β + ( 2 sin β) + ( 2 sin β)
r0 k sin θ k sin θ

EL i ( k r 0 − ω t) b sin( β) 4 sin β cos 2 α E b sin( β) 4 sin β cos 2 α
EP = e + = L e i ( k r 0 − ω t) +
r0 β k sin θ r0 β 2 β/ b
EL b i ( k r 0 − ω t ) 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 sin 3 α sin 3 α
1 + 2 cos 2 α = 1 + 2 1 − 2 sin 2 α = 3 − 4 sin 2 α = =
sin α sin α
So,
2 2
sin β sin 3 α
I = I0
β sin α
A plot is shown on the next page.
60
1 1 -1 8. 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
1 1 -1 9. 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
1 1 -2 0. The irradiance is given by,

2 2
sin β sin N α
I = I0
β α
π
with N = 1 0, a = 5 b , b = 1 0 − 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
1 1 -2 1 . Consider an N slit aperture, with N an odd number, arranged like the 3 slit aperture of problem 1 1 1 8
with the origin at the middle of the central slit. Then, the electric field at point P on the screen has the
form,
( Zb − 1 )/2
( NX
" Z b Z ja + b #)
− ja + 2
EL i ( k r 0 − ω t) 2
i k s sin θ i k s sin θ
2
i k s sin 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,
Zb
2 b sin β k
e i k s s in θ ds = , 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 − 2 i jα ( e i β − e − i β ) + e − 2 i jα ( e i β − e − i β ) 4 i ( sin β) ( cos 2 j α)
e i k s sin θ ds + e i k s sin θ ds = =
b
− ja − 2 ja − 2
b i k sin θ i k sin θ
Z − ja + b Z ja + b
2 2 2 b ( sin β) ( cos 2 j α)
e i k s sin θ ds + e i k s sin θ ds =
− ja −
b
ja −
b β
2 2
k
where α = a sin θ. Then,
2
− 1 )/2
( NX
" Z b Z ja + b # − 1 )/2
( NX
− ja + 2
i k s s in θ
2
i k s sin θ 2 b ( sin β)
e ds + e ds = cos 2 jα
− ja −
b
ja −
b β
j= 1 2 2 j= 1
The sum can be performed as,
− 1 )/2
( NX − 1)/2
( NX
! ( − 1)/2
( NX
) ) (
2iα (N − 1 )/2
( e ) − 1
cos 2 jα = Re e 2 i jα = Re ( e 2 i α ) j = Re
e2iα − 1
j=1 j= 1 j= 1
− 1)/2
( NX i Nα
e − e− iα cos Nα + i sin Nα − cos α + i sin α sin Nα 1
cos 2 jα = Re = Re = −
eiα − e− iα 2 i sin α 2 sin α 2
j= 1
− 1 )/2
( NX
" Z b Z ja + b # (N − 1 )/2
− ja + 2
i k s sin θ
2
i k s sin θ 2 b ( sin β) X b ( sin β) sin Nα
e ds + e ds = cos 2 jα = −1
− ja −
b
ja −
b β β sin α
j= 1 2 2 j= 1

EL b i ( k r 0 − ω t) sin β sin Nα E b sin β sin Nα
EP = e 1− + 1 = L e i ( k r0 − ω t )
r0 β sin α r0 β sin α
2 2
sin β sin Nα
I = I0
β sin α
1 1 -2 2 . ( a) See Figure 1 1 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 . 1 51 / mm) y
2 λ f ( 546 × 1 0 − 6 ) ( 1 000) mm
π ( 0. 1 ) x
β= mm − 1 = ( 0. 575/ mm) x
( 546 × 1 0 − 6 ) ( 1 000)
2 2
sin ( 1 . 1 51 y/ mm) sin ( 0. 575 x/ mm) sin 2 ( 1 . 1 51 y/ mm) sin 2 ( 0. 575 x/ mm)
I = I0 = I0
1 . 1 51 y/ mm 0. 575 x/ mm ( 0. 438 x 2 y 2 ) / mm 2
( 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 . 1 51 ) 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 . 1 51 0. 575
2 2
sin ( 3 × 1 . 1 51 ) sin ( 2 × 0. 575 )
( d) At x = 2 mm, y = 3 mm, I0 = 0. 005 I0
3 × 1 . 1 51 2 × 0. 575
62
1 1 -2 3. Since m λ = a sin θ , with m = 1 , sin θ 1 / 2 = λ/ a.
( a) sin θ 1 / 2 = λ/ λ = 1 ⇒ θ 1 / 2 = 90 ◦ ( b) sin θ 1 / 2 = λ/ 5 λ = 0. 2 ⇒ θ 1 / 2 = 1 1 . 5 ◦
( c) sin θ 1 / 2 = λ/ 1 0λ = 0. 1 ⇒ θ 1 / 2 = 5. 7 ◦
1 1 -2 4. Using the large γ expression,

2
J1 ( γ) I sin γ − cos γ
I = I0 ≈ 02 √
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
1 1 -2 5. We have I = I0 ( sin 2 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
I0
= 0. 0450 I0 sin( 1 . 43 π)
I1 ≈ I1 = I0 = 0. 0472 I0
( 1 . 5 π) 2 1 . 43 π
2
I0
= 0. 01 62 I0 sin( 2. 46 π)
I2 ≈ I2 = I0 = 0. 01 65I0
( 2. 5 π) 2 2. 46 π
2
I0
= 0. 00827 I0 sin( 3. 47 π)
I3 ≈ I3 = I0 = 0. 00834 I0
( 3. 5 π) 2 3. 47 π
1 1 -2 6. The principal maxima satisfy m λ = a sin θ, so that,
a 2/ 3 2
m m 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
λ 3 1
sin θ = ± = ± ⇒ θ = 30 ◦
3 2 2
The angular half width of the beam is 30 ◦ .
63
1 1 -2 7. 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 = 1 20 ◦ . Or, using Eq. ( 1 6 33) , I = 0 when α = pπ/ N
with p = ± 1 , ± 2, . . .
π 2π 2π 4π
Then α = ± , ± , . . . and since ϕ = 2 α, ϕ = ± ,± , . . . = ± 1 20 ◦ , ± 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, Ip ∝ 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 / Im ax = 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 .
1 1 -2 8. 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 = 1 80 ◦ ϕ = 270 ◦
64
C hapter 1 2 The Diffraction G rating
1 2 -1 . The angular positions of constructive interference are described by m λ = a sin θ. For the case at hand m =
2, λ 1 = 400 nm, and λ 2 = 600 nm. Then,

m λ1 2 ( 400 × 1 0 − 7 ) cm
θ 1 = sin − 1 = sin − 1 = 23. 58 ◦
a 1 / 5000 cm

− 1 m λ2 2 × 600 − 7 cm
θ 2 = sin = sin −1
= 36. 87 ◦
a 1 / 5000 cm
The angular separation is then,
0
∆ θ = θ 2 − θ 1 = 1 3. 292 ◦ = 1 3 ◦ 1 8
1 2 -2 . ( a) See Figure 1 2 1 3 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 = 1 4366 rad/ cm
d λ a cos θ a cos ( sin − 1 m λ/ a) cos sin − 1
3 × 650 × 1 0 − 7
3500 1 /3500
rad 1 80 deg cm
D = 1 4366 × × 1 0− 7 = 0. 0823 deg/ nm
cm π rad nm
The linear dispersion is then,

dy dθ − 7 rad
=f = f D = 1 500 mm 1 4366 × 1 0 = 2. 1 549 mm/ nm
dλ dλ nm
So the plate factor is,

dλ
= 0. 464 nm/ mm
dy
( b) According to Eq. ( 1 2 1 1 ) he resolving power is, R = m N = ( 3 ) ( 3500/ cm) ( 6 cm) = 63, 000
1 2 -3. ( a) Principal maxima: m λ = a sin θ. So, for the maximum

mλ 2 × 600 × 1 0 − 7
θ 1 = sin −1
= sin −1
= 6. 8921 ◦
a 1 0− 3
1
Minima: m ± λ = a sin θ. So for the neighboring minimum,
N
1 1
m± N λ 2 + 2 4 600 × 1 0 − 7
θ 2 = sin − 1 = sin − 1 = 7. 0364 ◦
a 1 0− 3

1
Then, ∆ θ = θ 2 − θ 1 = 0. 1 443 ◦ = 8. 7 0 . For m − , we get sin θ = 0. 1 1 75, θ = 6. 7478 ◦ and ∆ θ = 8. 7 0 again.
N
( b) At the minimum, we found θ = 7. 0364 ◦ or sin θ = 0. 1 225. Thus,
a sin θ ( 1 0 − 3 c m ( 0. 1 225)
λ= = = 6. 1 25 × 1 0 − 5 cm = 61 2. 5 nm
m 2

1
If we had taken m− = ( 2 − 1 / 24) instead, sin θ = 0. 1 1 75 and
N
a sin θ ( 1 0 − 3 cm) ( 0. 1 1 75)
λ= = = 5. 875 × 1 0 5 cm = 587. 5 nm
m 2
( c) The minimum angular separation is ( ∆ λ) m in = 61 2. 5 nm − 600 nm = 1 2. 5 nm. So,
λ 600
Rm = 2 = = = 48
∆ λ m in 1 2. 5
Using Eq. ( 1 2 1 1 ) , R = m N = ( 2) ( 24) = 48.
65
λ a v 589. 2935 a
1 2 -4. R = m N = = = 987 ; N = . So, for m = 1 , N = 987 and for m = 2, N = 494.
∆λ 0. 597 m
1 2 -5. The free spectral range is FSR = λ 2 − λ 1 = λ 1 / m. Let λ 1 = 350 nm, crown ; λ 1 = 1 80 nm, quartz.
( a) Then,
λ1 350 nm
λ2 = λ1 + = 350 nm + = 700 nm, crown
m 1
1 80 nm
λ 2 = 1 80 nm + = 360 nm, quartz
1
( b) With a = 1 / 1 200 mm,

( 1 ) ( 700 × 1 0 − 6 )
θ = sin −1
( m λ/ a) = sin −1
= 57. 1 4 ◦ , crown
1 / 1 200

( 1 ) ( 360 × 1 0 − 6 )
θ = sin − 1 ( m λ/ a) = sin − 1 = 25. 59 ◦ , quartz
1 / 1 200
( c) FSR = λ 1 / m:
m = 1 : FSR = 350/ 1 = 350 nm, crown; FSR = 1 80/ 1 = 1 80 nm, quartz

m = 2: FSR = 350/ 2 = 1 75 nm, crown; FSR = 1 80/ 2 = 90 nm, quartz
1 2 -6. See Figure 1 2 1 4 that accompanies the statement of this problem in the text.
R ( m = 3 ) = m N = ( 3 ) ( 1 6, 000 × 2. 5) = 1 20, 000

R ( m = 2) = m N = ( 2) ( 1 6, 000 × 2. 5) = 80, 000
∆ λ = λ/ R = 550 nm/ 80, 000 = 0. 006875 nm = 0. 069 A ˚
λ 5893
1 2 -7. ( a) R ≡ = = 982. 1 6 ⇒ 983, m = R/ N = 983/ 400 = 2. 46; order = 3
∆λ 6
( b) Any width smaller than the light beam.
1 2 -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 × 1 0 − 6 ) ( 2000)
m = 1 : y1 = mm = 21 8. 4 mm
0. 005
Thus ∆ y = y1 − y0 = 21 . 84 cm , 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 × 1 0 − 7 )
N = 2: ∆ y = cm = 21 . 84 cm, N = 1 0: ∆ y = 4. 368 cm, N = 1 5, 000: ∆ y = 0. 0029 cm
( 0. 0005) ( 2)
66
1 2 -9. ( a) The resolution is R = λ/ ∆ λ = 3500 A ˚ / 0. 02 A
˚ = 1 75, 000 = m N. So for m = 2, N = 87, 500 grooves so that
there are 87, 500/ 1 0 = 8750 grooves/ cm.
( b) At normal incidence,

1 mλ 1 2 × 350 × 1 0 − 7 cm
θ B = sin − 1 sin − 1 = 1 8. 89 ◦
2 a 2 1 / 8750 cm
( c) m 2 rad
( d) D = = = 221 39
a cos θ ( 1 / 8750) cos 37. 77 cm
θb rad 1 80 deg cm
θb
θ D = 221 39 × × 1 0− 7 = 0. 1 268 deg/ nm
θb cm π rad nm
dλ 1
= = 7. 88 nm/ deg
θ = 2 θ b = 37. 77 ◦ dθ D
1 2-1 0. ( a) The resolution is given by R = λ/ ∆λ m in . Taking the worst case, or λ = 700 nm,
˚
7000 A
R= = 7000 = m N = ( 1 ) N ⇒ N = 7000
˚
1A
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 × 1 0 − 8 cm)
∆ y= = = 0. 001 75 cm
a ( 2/ 7000) cm
mλ
1 2-1 1 . ( a) Using m λ = a ( sin θ i = sin θ m ) ⇒ sin θ m = − sin θ i ,
a

( 1 ) ( 400)
λ = 400 nm: θ m = sin − 1 − sin 30 = − 5. 74 ◦
◦
1 000

( 1 ) ( 700)
λ = 700 nm: θ m = sin − 1 − sin 30 ◦ = + 1 1 . 54 ◦
1 000
( b) W = 1 0 cm: R = m N = ( 1 ) ( 1 0, 000 cm − 1 ) ( 1 0 cm) = 1 00, 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/ 1 000) )

D = tan θ/ λ = = = 0. 001 1 974 rad/ nm
λ 550 nm
So the plate factor is
dλ 1 ˚ / mm
= = 8. 351 6 nm/ cm = 8. 35 A
d y ( 1 00 cm) ( 0. 001 1 9737) nm − 1
( d) R c irc le = R gra t in g / 2 = ( 1 / 2) ( 2 m) = 1 m
1 2-1 2 . 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 × 1 06 = 5000 grooves/ cm
a ( 1 ) ( 1 00 cm) ( 2 nm/ mm) nm − cm mm
67
1 2-1 3. ( a) For the Littrow mount: m λ = 2 a sin θ b , and so,
2 a sin θ b
λ= = 2 ( mm/ 300) sin 1 0 ◦ = 0. 001 1 6 mm = 1 . 1 6 µm.
m
( b) See the solution to problem 1 2 1 2. In this case,
a cosθ ( 1 mm/ 300) ( cos 20. 322 ◦ ) ( 1 0 6 nm/ cm)
P. F. = = = 1 . 84 nm/ mm
mf ( 1 ) ( 1 70 cm) ( 1 0 mm/ cm)
( Here, sin θ = m λ/ a = 2 sin θ b = 2 sin 1 0 ◦ ⇒ cos θ = cos 20. 322 ◦ )
1 2-1 4. See Figure 1 2 1 6 that accompanies the statement of this problem in the text. Generally,
R = m N = m ( W/ a) = ( 1 ) ( 1 5 cm/ a) or a = ( 1 5 cm) / R = ( 1 5 cm) / 300, 000 = 5 × 1 0 − 5 cm = 500 nm

mλ 1 × 200
( a) Littrow: θ b = sin − 1 = sin − 1 = 1 1 . 5◦
2a 2 × 500

mλ 1 × 200
( b) Normal: 2 θ b = sin −1
= sin −1
= 23. 578 ◦ . So that, θ b = 1 1 . 8 ◦
a 500
1 2-1 5. As given in the statement of the problem,

λ 488 × 1 0 − 6 mm 1
d= = = 2. 81 75 × 1 0 − 4 mm ⇒ = 3550 grooves/ mm
2 sin θ 2 sin 60 ◦ d
If the emulsion is of refractive index n, then λ → λ 0 / n and d = λ 0 / ( 2 n sin θ) . As n increases the grating con
stant increases.
λ 6563 6563
1 2-1 6. ( a) R = = = m N or N = = 3647 grooves
∆ λ m in 1.8 1.8 (1)
!
m λ 1 × 6300 ˚
A ˚
6300 A
( b) Littrow: θ b = sin − 1 or 22. 2 ◦ = sin − 1 ⇒ = 0. 37784 ⇒ a = 8. 337 × 1 0 − 5 cm.
2a 2a 2a
1
Thus, = 1 1 995 grooves/ cm ≈ 1 200 grooves/ mm.
a
3647 gr
( c) Wm in = = 3. 04 mm
1 200 gr/ mm
1 2-1 7. ( a) Littrow: m λ = 2 a sin θ b ,

2 a sin θ b 2 ( 1 cm/ 80) sin 63 ◦ sin 63 ◦ cm
So m = = = .
λ λ 40 λ
λ = 400 × 1 0 − 7 cm, m = 557
λ = 700 × 1 0 − 7 cm, m = 31 8
W
( b) N = = ( 1 2 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 × 1 0 − 7
λ ˚
5500 A
Then, ( ∆ λ) m in = = ˚
= 0. 01 4 A
R 388, 800
m 405 rad 1 80 deg cm
( d) D = = = 71 367 × × 1 0− 7 = 0. 41 deg/ nm.
a cos θ c m cos 63 ◦ cm π rad nm
80

λ m in 350 sin 63 1 350 nm
( e) FSR = = , where m = = 636. 4. So, FSR = = 0. 55 nm = 5. 5 ˚A .
m m 40 350 × 1 0 − 7 637
68
C hapter 1 3 Fresnel Diffraction
1 3-1 . The far field is defined by

Area π d 2 π( 1 0 − 3 m) 2
L = = = 1.4 m
λ 4λ 4 ( 546 × 1 0 − 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.
1 3-2 . See Figure 1 3 9 that accompanies the statement of the problem in the text. From Eq. ( 1 3 23) ,
R 12 f1
fn = =
nλ n
Then, the maxima occur for odd values of n at,
( 0. 1 5 cm) 2 f f
f1 = = 409 cm, f3 = 1 = 1 36 cm, f5 = 1 = 81 . 8 cm
( 1 ) ( 5. 5 × 1 0 − 5 cm) 3 5
The minima occur at even values of n,
f1 f f
f2 = = 204. 5 cm, f4 = 1 = 1 02 cm, f6 = 1 = 68. 1 cm
2 4 6
1 3-3. For a fixed distance f, we have the relation for the maxima and minima, f = r 0 = R n2 / nλ, with the maxima
occurring for odd n and the minima occurring for even n. Then the hole diameters are, then
√
D n = 2 R n = 2 n λ r0
p
First Max: n = 1 : D 1 = 2 ( 1 ) ( 5. 893 × 1 0 − 5 cm) ( 1 50 cm) = 0. 1 88 cm
p
Second Max: n = 3 : D 3 = 2 ( 3 ) ( 5. 893 × 1 0 − 5 cm) ( 1 50 cm) = 0. 326 cm
p
First Min: n = 2: D 2 = 2 ( 2) ( 5. 893 × 1 0 − 5 cm) ( 1 50 cm) = 0. 266 cm
p
Second Min: n = 4 D 4 = 2 ( 4) ( 5. 893 × 1 0 − 5 cm) ( 1 50 cm) = 0. 376 cm
√ q
1 3-4. ( a) R 1 = n λ r0 = ( 1 ) 6 × 1 0 − 5 cm) ( 20 cm) = 0. 0346 cm
( b) R 1 = 1 cm; N = R 12 / λ r 0 = 1 2 / ( 6 × 1 0 − 5 × 20) = 833
( c) No minima are appear now. Maxima are

R 12 ( 0. 0346 cm) 2 f f
f1 = r 0 = = = 20 cm; f3 = 1 = 6. 67 cm; f5 = 1 = 4 cm
( 1 ) λ ( 6 × 1 0 − 5 cm) 3 5
1 3-5. See Figure 1 3 20 that accompanies this problem in the text for the geometry of the set up. Let d 1 be the
distance from the source point to the indicated point on the wavefront in the aperture and d 2 be the dis
tance from the indicated point on the wavefront in the aperture to the observation point P. For the n t h
zone,
( d 1 + d 2 ) − ( p + q) = nλ/ 2
From the geometry,
p 1/2
R2 1 R n2 1 R 2n
d1 = p2 + R 2n = p 1 + 2n ≈p 1+ ; d 2 ≈ q 1 +
p 2 p2 2 q2
So the condition becomes,

1 1 1 1 1
( d 1 + d 2 ) − ( p + q) = R 2n + ≡ R n2 = nλ/ 2
2 p q 2 L
√
Rn = n λ L
69
1 3-6. Using the set up of problem 1 3 5,
1 1 1 1 1
= + = + ⇒ L = ( 20/ 3 ) cm
L p q 1 0 cm 20 cm
√ p
( a) R 1 = nλL= ( 1 ) ( 20/ 3 ) ( 6 × 1 0 − 5 ) cm = 0. 2 mm
R 21 (1)2
( b) N = = = 2500.
λ L ( 20/ 3 ) ( 6 × 1 0 − 5 )
1 3-7. Using the results of problem 1 3 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)
n in n er = =2
( 25 cm) ( 5 × 1 0 − 5 cm)
2
( 0. 0935 cm)
n ou t er = =7
( 25 cm) ( 5 × 1 0 − 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 ∝ ( a 1 / 2) 2 and here I ∝ a 21 assuming a 1 ≈ a 2 ≈ . . . a 7 . So,
I/ Iu = 4.
( b) In this case,
2
( 0. 1 cm)
n ou t er = =8
( 25 cm) ( 5 × 1 0 − 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) .
1 3-8. From Eq. ( 1 3 1 5) , the surface area of the n t h Fresnel zone is

2!
π r 00 r02 λ λ
Sn = + (2 n − 1)
r 0 + r00 r0 2 r0
So that, with λ/ r0 = ( 5 × 1 0 − 5 / 50) = 1 0 − 6

S2 5 − S1 48 ( λ/ 2 r0 ) 2
= = 1 . 2 × 1 0 − 5 = 0. 001 2%
S1 ( λ/ r 0 ) + ( λ/ 2r 0 ) 2
1 3-9. ( a) Before reduction R 1 = 1 1 . 25 cm.

√ p
The required radius is R 1 = n λ f1 = ( 632. 8 × 1 0 − 5 ) ( 200) = 0. 1 1 25 cm.
It must be reduced by a factor of 1 00.

√ √
( b) R N = N R1 = 20 ( 1 1 . 25 cm) = 50. 31 cm.
√
1 3-1 0. See Figure 1 3 21 that accompanies the problem in the text. Using R n = n r 0 λ ,
√ p
D 2 = 2 R 2 = 2 2 r 0 λ = 2 2 ( 25) ( 5. 5 × 1 0 − 5 ) cm = 0. 1 05 cm
√ p
D 4 = 2 R 2 = 2 4 r 0 λ = 2 4( 25) ( 5. 5 × 1 0 − 5 ) cm = 0. 1 48 cm
√ p
D 6 = 2 R 2 = 2 6 r 0 λ = 2 6 ( 25) ( 5. 5 × 1 0 − 5 ) cm = 0. 1 82 cm
70
1 3-1 1 . The zone areas are given by Eq. ( 1 3 1 5) which goes to Eq. ( 1 3 1 6) in the limit λ/ r 0 = λ/ x 1 . Eq. ( 1 3 1 6)
is independent of n, so the areas are approximately constant. For a plane wave, r 0 → ∞ , thus from Eq. ( 1 3
1 6) ,
π r00 r0 πλ
Sn = πλ = Lim = π λ r 00 = π λ x
r0 + r 00 r0 → ∞ 1
+
1
r0 r 00
√ p
1 3-1 2 . The radius is given by R 4 = 4 r 0 λ = ( 4 ) ( 2) ( 485 × 1 0 − 9 ) m = 0. 001 97 m = 1 . 97 mm. Since 4 zones con
tribute and the zones nearly cancel in pairs, the irradiance is nearly zero.
1 3-1 3. Since p → ∞ ,
1 1 1 1
= + = ⇒L= q
L p q q
Also ∆ z = w. Using Eq. ( 1 3 28) ,
r
2 2 w2 2 ( 0. 0005 m) 2
∆v = ∆z ⇒ L = = = 0. 1 48 m = 1 4. 8 cm
Lλ ( ∆ v) λ ( 2. 5) 2 ( 540 × 1 0 − 9 m)
1 3-1 4. See Figure 1 3 22 that accompanies the statement of this problem in the text. Also it may be helpful to
examine Figure 1 3 1 5 in the body of the text. The parameter L is,
1 1 1 1 1
= + = + ⇒ L = 1 0 cm
L p q 30 cm 1 5 cm
( a) For an on axis point, z2 = w/ 2 = 2. 5 × 1 0 − 4 m and z 1 = − w/ 2 = − 2. 5 × 1 0 − 4 m. Then

r r
2 2
v2 = z2 = ( 2. 5 × 1 0 − 4 ) = 1 . 694
Lλ ( 0. 1 ) ( 435. 8 × 1 0 − 9 )
r r
2 2
v1 = z1 = ( − 2. 5 × 1 0 − 4 ) = − 1 . 694
Lλ ( 0. 1 ) ( 435. 8 × 1 0 − 9 )
Then, interpolating in Table 1 3 1 ,
C( v 2 ) ≈ 0. 3265, S( v 2 ) = 0. 5549
C( v 1 ) = − 0. 3265, S( v 1 ) = − 0. 5549
Using Eq. ( 1 3 35) ,

n o
I = I0 [ C( v 2 ) − C( v 1 ) ] 2 + [ S( v 2 ) − S( v 1 ) ] 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 z 2 = 0. Then the bottom edge of the aperture is located at
z = − w = − 5 × 1 0 − 4 m. Then,
r r
2 2
v 2 = 0, v 1 = z1 = ( − 5 × 1 0 − 4 ) = − 3. 388
Lλ ( 0. 1 ) ( 435. 8 × 1 0 − 9 )
C( v 2 ) = 0, S( v 2 ) = 0; C( v 1 ) = − 0. 42746, S( v 1 ) = − 0. 49328
n o
2 2
I = I0 [ C( v 2 ) − C( v 1 ) ] + [ S( v 2 ) − S( v 1 ) ]
I = I0 0. 4276 2 + 0. 49328 2 ) = 0. 4262 I0 = 0. 4262 ( Iu / 2) = 0. 21 31 Iu
71
1 3-1 5. Refer to Figure 1 3 23 that accompanies the statement of the problem in the text and Figure 1 3 4a in the
body of the text. For both positions,
1 1 1
= + ⇒ L = 40 cm = 0. 4 m
L 60 cm 1 20 cm
( a) Let z 0 be the coordinate of the point O 0 in the aperture plane along the straight line from S to the
observation point P. ( See Figure 1 3 4a) . In the case at hand
p 60
z0 = y= ( − 2 mm) = − 6. 667 × 1 0 − 4 m.
p+ q 60 + 1 20
The values of the z 1 and z2 that mark the edges of the unobstructed regions in the aperture plane are to be
measured relative to this point so that z 2 = ∞ and z 1 = + 6. 667 × 1 0 − 4 m
r r
2 2
v2 = ∞ , v 1 = z1 = 6. 667 × 1 0 − 4 = 1 . 942
λL ( 589. 3 × 1 0 ) ( 0. 4)
− 9
C( v 2 ) = 0. 5, S( v 2 ) = 0. 5, C( v 1 ) = 0. 4338, S( v 1 ) = 0. 3608
n o
I = I0 [ C( v 2 ) − C( v 1 ) ] 2 + [ S( v 2 ) − S( v 1 ) ] 2

I = I0 0. 5 − 0. 4338) 2 + ( 0. 5 − 0. 3608) 2 = 0. 02376 I0 = 0. 02376 ( Iu / 2) = 0. 01 1 9 Iu
p 60
( b) In this case z 0 = y= ( 1 mm) = 3. 333 × 1 0 − 4 m. Then, z 2 = ∞ and z 1 = − 3. 333 × 1 0 − 4 m
p+ q 60 + 1 20
r r
2 2
v 2 = ∞ , v1 = z1 = − 3. 333 × 1 0 − 4 = − 0. 9709
λL ( 589. 3 × 1 0 ) ( 0. 4)
− 9
C( v 2 ) = 0. 5, S( v 2 ) = 0. 5, C( v 1 ) = − 0. 7755, S( v 1 ) = − 0. 4097
n o
I = I0 [ C( v 2 ) − C( v 1 ) ] 2 + [ S( v 2 ) − S( v 1 ) ] 2

I = I0 0. 5 + 0. 7755) 2 + ( 0. 5 + 0. 4097) 2 = 2. 454 I0 = 2. 454 ( Iu / 2) = 1 . 23 Iu
1 3-1 6. 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 z u2 = ∞ and
z u 1 = 7. 5 × 1 0 − 4 m. The unobstructed region in the lower half of the aperture plane is bounded by the
points z l 2 = − 7. 5 × 1 0 − 4 m and z l 1 = − ∞ . Then,
r r
2 2
v u2 = ∞ , v u1 = z1 = 7. 5 × 1 0 − 4 = 3. 2094
λL ( 546. 1 × 1 0 ) ( 0. 2)
− 9
C( v u 2 ) = 0. 5, S( v u2 ) = 0. 5, C( v u 1 ) = 0. 4607, S( v u1 ) = 0. 5863
v l 2 = − 3. 2094, v l 1 = − ∞
C( v l 2 ) = − 0. 4607, S( v l 2 ) = − 0. 5863, C( v l 1 ) = − 0. 5, S( v l 1 ) = − 0. 5
n o
2 2
I = I0 [ C( v l 2 ) − C( v l 1 ) + C( v u2 ) − C( v u1 ) ] + [ S( v l 2 ) − S( v l 1 ) + S( v u 2 ) − S( v u 1 ) ]
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. 01 8 Iu
Part ( b) is on the next page.
72
1 3-1 6. ( b) The point on the screen that is at the edge of the geometrical shadow is at position
p+ q 90
y= z w ire t op = ( 0. 75 mm) = 2. 25 mm
p 30
Following the procedure from part ( a) :
z u 1 = 0, z u2 = ∞ , z l 1 = − ∞ , z l 2 = − 1 . 5 × 1 0 − 3 m
r
2
vu 1 = 0, v u 2 = ∞ , v l 1 = − ∞ , v l 2 = − 1 . 5 × 1 0 − 3 = − 6. 42
( 546. 1 × 1 0 ) ( 0. 2)
− 9
C( v u2 ) = 0. 5, S( v u2 ) = 0. 5, C( v u 1 ) = 0, S( v u1 ) = 0
n C( v l 2 ) = − 0. 541 4 , S( v l 2 ) = − 0. 51 38, C( v l 1 ) = − 0. 5, S( v l 1 ) = − 0. 5 o
I = I0 [ C( v l 2 ) − C( v l 1 ) + C( v u2 ) − C( v u1 ) ] 2 + [ S( v l 2 ) − S( v l 1 ) + S( v u 2 ) − S( v u 1 ) ] 2
n o I
I = I0 [ − 0. 541 4 − ( − 0. 5) + 0. 5 − 0] 2 + [ − 0. 51 38 − ( − 0. 5) + 0. 5 − 0] 2 = 0. 4467I0 = 0. 4467 u = 0. 223 Iu
2
1 3-1 7. Refer to the routine used to solve problem 1 3 1 5b. Borrowing from that solution,
n o n o
I = I0 [ C( v 2 ) − C( v 1 ) ] 2 + [ S( v 2 ) − S( v 1 ) ] 2 = I0 [ 0. 5 − C( v 1 ) ] 2 + [ 0. 5 − S( v 1 ) ] 2
From Table 1 3 1 the second maximum and minimum occur for,
Max: v 1 = − 2. 3, C( v 1 ) = − 0. 626, S( v 1 ) = − 0. 5531
I2 n d m ax = 2. 378 I0 = 2. 378 ( Iu / 2) = 1 . 1 9 Iu
Min: v 1 = − 2. 75, C( v 1 ) = − 0. 4300, S( v 1 ) = − 0. 4222
I2 n d m in = 1 . 71 5 I0 = 1 . 71 5 ( Iu / 2) = 0. 86 Iu
1 1 1
1 3-1 8. 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 z u2 = ∞ and z u1 = 1 . 85 × 1 0 − 4 m. The unobstructed region in
the lower half of the aperture plane is bounded by the points z l 2 = − 1 . 85 × 1 0 − 4 m and z l 1 = − ∞ . Then,
r r
2 2
vu 2 = ∞ , vu 1 = z1 = 1 . 85 × 1 0 − 4 = 0. 3009
λL ( 630 × 1 0 ) ( 1 . 2)
− 9
C( v u 2 ) = 0. 5, S( v u2 ) = 0. 5, C( v u 1 ) = 0. 3003, S( v u1 ) = 0. 01 43
v l 2 = − 3. 2094, v l 1 = − ∞
n C( v l2 ) = − 0. 3003, S( v l 2 ) = − 0. 01 43, C( v l 1 ) = − 0. 5, S( v l 1 ) = − 0. 5 o
I = I0 [ C( v l 2 ) − C( v l 1 ) + C( v u2 ) − C( v u1 ) ] 2 + [ S( v l 2 ) − S( v l 1 ) + S( v u 2 ) − S( v u 1 ) ] 2
nh o
I = I0 − 0. 3003 − ( − 0. 5) + 0. 5 − 0. 3003] 2 + [ − 0. 01 43 − ( − 0. 5) + 0. 5 − 0. 01 43] 2 = 1 . 1 03I0
I = 1 . 1 03 ( Iu / 2) = 0. 55Iu
1 3-1 9. See Figure 1 3 25 that accompanies the statement of the problem in the text. The parameter L is given by
1 1 1
= + ⇒ L = 5 cm
L 1 0 cm 1 0 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,
z 1 = ( 0. 25/ 2 − 0. 02) cm = 0. 01 05 cm, z 2 = ( 0. 25/ 2 + 0. 02) cm = 0. 01 45 cm
r r
2 2
v1 = z1 = ( 0. 01 05 ) = 0. 8987, v 2 = 1 . 241 1
λL 5 ( 5. 46 × 1 0 − 5 )
n C( v 1 ) = 0. 76426, S( v 1 ) = 0. 33862 , C( v 2 ) =o 0. 68384, S( v 2 ) = 0. 64925
I = I0 [ 2 ( C( v 2 ) − C( v 1 ) ) ] 2 + [ 2 ( S( v 2 ) − S( v 1 ) ) ] 2 = 0. 41 2 I0 = 0. 41 2 Iu / 2 = 0. 206 Iu
73
1 3-2 0. The parameter L is given by
1 1 1
= + ⇒ L = 1 2. 5 cm
L 25 cm 25 cm
( a) The far field condition is

p, q b 2 / λ = ( 0. 075 cm) 2 / ( 4. 358 × 1 0 − 5 cm = 1 29 cm
Since for the case described in the problem statement p = q = 25 cm, 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 × 1 0 − 5 ) / 0. 075 ⇒ θ = 0. 03329 ◦
y = q tan θ = ( 25 cm) tan ( 0. 00329 ◦ ) = 0. 01 45 cm
( c) See Figure 1 3 1 5a for the geometry. In the case at hand, y = 0. 01 45 cm so that the point O 0 in the aper
ture plane and connecting the source to the observation point is positioned at,
p 1
z0 = 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. 71 1 5
Lλ ( 1 2. 5) ( 4. 358 × 1 0 − 5 )
r r
2 2
v2 = z1 = ( 0. 03025) = 1 . 8329
Lλ ( 1 2. 5) ( 4. 358 × 1 0 − 5 )
C( v 1 ) = − 0. 401 , S( v 1 ) = − 0. 446; C( v 2 ) = 0. 354, S( v 1 ) = 0. 425
n o
2 2
I = I0 [ C( v 2 ) − C( v 1 ) ] + [ S( v 2 ) − S( v 1 ) ]
n o
I = I0 ( 0. 354 + 0. 401 ) 2 + ( 0. 425 + 0. 446) 2 = 1 . 33I0 = 1 . 33( Iu / 2) = 0. 665 Iu
1 3-2 1 . 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 × 1 0 − 9 m)
a= = ≈ 1 9 µm
sin θ sin ( 2 ◦ )
74
C hapter 1 4 Matrix Treatment of P olarization
1 4-1 . 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 ◦ ) a b sin θ 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 sin 2 θ + cos 2 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 = cos 2 θ , d = sin 2 θ
So that,

a b sin 2 θ sin θ cos θ
=
c d sin θ cos θ cos 2 θ

E0 x e i ϕ x i ( k z − ωt)
1 4-2 . In general Ẽ = E0 x e i ϕ x xx + E0 y e i ϕ y ŷ e i ( k z − ωt) = e = Ẽ0
E0 y e i ϕ y

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

1 1 1
( b) Ẽ = [ E0 x̂ + E0 ŷ ] e i ( k z − ωt) ⇒ Ẽ0 = E0 ⇒√ . Linearly polarized at 45 ◦ .
1 2 1
 
h i
1 1 1
( c) Ẽ = E0 x̂ + E0 e − i π/ 4 ŷ e i ( k z − ωt) ⇒ Ẽ0 = E0 − i π/ 4 ⇒ √  1  . Then,

e 2 √ ( 1 − i)
2
2E0 x E0 y cos ε
tan 2α = → ∞ ⇒ 2 α = 90 ◦ , α = 45 ◦
E02x − E02 y
Right elliptically polarized at 45 ◦ .
h i
1 1 1
( d) Ẽ = E0 x̂ + E0 e i π/ 2 ŷ e i ( k z − ωt ) ⇒ Ẽ0 = E0 ⇒√ . Left circularly polarized.
e i π/ 2 2 i
75

E0 x e i ϕ x i ( k z − ωt )
1 4-3. In general Ẽ = E0 x e i ϕx
x̂ + E0 y e iϕy
yy e i ( k z − ωt )
= e = Ẽ0 e i ( k z − ωt)
E0 y e i ϕ y

1
( a) Ẽ = ( 2 E0 x̂ + 0 ŷ ) e i ( k z − ωt) ⇒ Ẽ0 = 2 E0 . Linearly polarized along the x direction. Velocity is in the
0
√
+ z direction. The amplitude is A = 2 E0 1 2 + 0 2 = 2 E0 .

3
( b) Ẽ = ( 3 E0 x̂ + 4E0 ŷ ) e i ( k z − 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 = 3 2 + 4 2 E0 = 5 E0 .

1
( c) Ẽ = 5 E0 ( x̂ − iŷ ) e i ( k z + ωt ) ⇒ Ẽ0 = 5 E0 . 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 ( k z − ωt) 1
1 4-4. ( a) Ẽ1 = E0 1 ( x̂ − ŷ ) e ⇒ Ẽ0 1 = 2 E0 1 . This is linearly polarized along − 45 ◦
−1
√ √
3 1
Ẽ2 = E0 2 3 x̂ + ŷ e i ( k z − ωt) ⇒ Ẽ0 2 = E0 2 . This is linearly polarized along α = tan − 1 √ = 30 ◦
1 3
The angle between the two is 75 ◦ . √
√ √ √ 3−1
( b) Ẽ0 1 · Ẽ0 2 = E0 1 E0 2 ( 3 − 1 ) = ( 2 E0 1 ) ( 3 + 1 E0 2 ) cos( θ 1 2 ) ⇒ cos θ 1 2 =
2 √ ⇒ θ 1 2 = 75 ◦
2 2
1 4-5. 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
H W P , FA vert L P at − 4 5 ◦
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
QWP LP
The output light has right circular polarization.
1 4-6. ( a) The wave propagates in the + x direction so the phase factor is e i ( k x − ω t) . The angle α = 30 ◦ so that
1 E √
tan α = 30 ◦ = √ = 0 z . Then Ẽ = E0 ( 3 ŷ + ẑ ) e i ( k x − ωt) .
3 E0 y
( b) The phase factor is e i ( k y − ω t ) . Also,

A 2
with A = 2 B ⇒ E0
− iB −i
So, Ẽ = E0 ( 2 ẑ − ix̂ ) e i ( k y − ωt )
| k| | k|
( c) The propagation vector is k = √ ( x̂ + ŷ ) . Then k · r = √ ( x + y) .
h i
2 2
√
i | k | ( x + y) / 2 − ωt
Then Ẽ = E0 ẑ e .
76
1 4-7. 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.
1 4-8. The form of the ellipse is given as,

2 2
Ex Ey Ex Ey
+ −2 cos ε = sin 2 ε
E0 x E0 y E0 x E0 y
This has the form of a quadratic,

E y2 + b E y + c = 0
with,
E0 y Ex2
b=− 2 ( cos ε) Ex and c = E02 y − sin 2 ε
E0 x E02x

A √
As shown in the text, when Ẽ0 = , E0 x = A, E0 y = 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:
( %i 1 )( A: 3, B: 2, C: 1 ) $
( %i 2) eps i lon: atan( C/B) $
( %i 3) ( E0x: A, E0y: s qrt( B^ 2 +C^ 2) ) $
( %i 4) f or j f rom - 30 thru 30 do
( Ex[ j ] : 0. 1 * j , b[ j ] : - 2* E0y/E0x* c os ( e ps i lon) * Ex[ j ] , c [ j ] : E0y^ 2* ( Ex[ j ] ^ 2/ E0x^ 2 - ( s i n( eps i lon) ) ^ 2) ,
Ey1 [ j ] : - b[ j ] /2 - 1 / 2* s qrt( b[ j ] ^ 2- 4* c [ j ] ) , Ey2[ j ] : - b[ j ] /2+1 / 2* s qrt( b[ j ] ^ 2- 4* c [ j ] ) ) $
( %i 5) f or j f rom 0 thru 0 do l di s pl ay( Ex[ j ] , Ey1 [ j ] , Ey2[ j ] )
( %t5) Ex0 = 0
( %t6) Ey1 0 = − 1
( %t7) Ey2 0 = 1
N ote that I have displayed the first set of p oints values. T he loops could b e extended to form any numb er of the values. T he
two E y values correspond to the two roots of the quadratic. T he list so generated can be plotted to form the ellipse. T his is
done below.
( %i 7) ( Eyl i s t: l i s tarray( Ey1 ) , Exli s t: l i s tarray( Ex) , Eyli s t2 : li s tarray( Ey2) ) $
( %i 8) pl ot2d( [ [ di s c rete , Exli s t, Eyli s t] , [ di s c rete, Exl i s t, Eyli s t2] ] , [ gnupl ot_te rm, ps ] , [ gnuplot_ps _term_ c ommand,
" s et te rm pos ts c ri pt e ps monoc hrome" ] , [ gnupl ot_out_ f i l e, " e ll i ps e. eps " ] )
25
Ey discrete1
discrete2
2
15
05
Ex
05
15
25
3 2 1 0 1 2 3
77

3i 3
1 4-9. ( a) →i Linear polarization at α = tan − 1 ( 1 / 3 ) = 1 8. 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 = 2 2 + 3 2 = 1 3
3

2 1 √
( g) =2 Left elliptical polarization, E0 x = 2, E0 y = 2 3 2 + 4 2 = 1 0, ε = tan − 1 ( 4/ 3 ) = 53. 1 ◦
6+8i 3 + 4i
( 2) ( 1 ) ( 5 ) cos 53. 1 ◦
tan 2 α = ⇒ α = − 7. 02 ◦
1 2 − 52
1 4-1 0. Using the Jones formalism, √ √

1 0 3 3
=
0 −i 1 −i
QWP LP, 30◦
√
The output light is has right elliptical polarization with E0 x / E0 y = 3.
1 4-1 1 . Using the Jones formalism,

1 0 cos α cos α cos( − α)
= =
0 −1 sin α − sin α sin( − α)
HWP LP, α LP, − α
The input light is rotated by 2 α.
1 4-1 2 . 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 1 4 1 1 . ) 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.
1 4-1 3. See Figure 1 4 1 3 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
QWP LP
S A , h or at 45◦
1 4-1 4. Using the Jones formalism,

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 h or at 4 5 ◦
1 4-1 5. Using the Jones formalism,

√
0 0 1 0 cos 30 ◦ 0 0 3 0 0
= = =−i
0 1 0 −i sin 30 ◦ 0 1 −i −i 1
LP QWP LP ( a ) R ight ( b) LP
TA ve r t S A h or at 30 ◦ E llipt vert
78
1 4-1 6. ( 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 RC P

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 RC P
( b) For an eighth wave plate: | ε y − ε x | = 2 π/ 8 = π/ 4. Then the Jones matrix is,

1 0
M1 / 8 =
0 e ± i π/ 4
with the ± for ε y > ε x or ε y < ε x . Then,

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

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

 
1
A √ √
= ± 2 ± i 2 
B+iC
2
√
This is of the form of elliptical polarization with A = 1 , and | B | = | C | = 2 / 2.
1 4-1 7. 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
1 4-1 8. Note that,

1 cos α cos α + 1 A
+ = =
±i sin α sin α ± i B±iC
C irc u lar L in ear E llip t ic al
79
1 4-1 9. In general, for elliptical polarization, E0 x does not equal E0 y and ε does not equal 0,
E y = E0 y cos( k z − ω t) = E0 y cos α
Ex = E0 x cos( k z − ω t + ε) = E0 x cos ( α + ε)
Eliminate z and t through α:
Ey
= cos α
E0 y
Ex E p
= cos ( α + ε) = cos α cos ε − sin α sin ε = y cos ε − 1 − ( E y / E0 y ) 2 sin ε
E0 x E
p 0y
( Ex / E0 x ) − ( E y / E0 y ) cos ε = − 1 − ( E y / E0 y ) 2 sin ε
Squaring gives,
( Ex / E0 x ) 2 + ( E y / E0 y ) 2 cos 2 ε − 2 ( Ex E y ) cos ε/ ( E0 x E0 y ) = ( 1 − ( E y / E0 y ) 2 sin 2 ε
2 2
Ex Ey E E
+ cos 2 ε + sin 2 ε − 2 x y cos ε = sin 2 ε
E0 x E0 y E0 x E0 y
2 2
Ex Ey E E
+ − 2 x y cos ε = sin 2 ε
E0 x E0 y E0 x E 0 y
1 4-2 0. ( a) Putting the Jones vector into standard form,

A 2 2 2 √
= = =
B+iC 3 e i π/ 3 3 cos ( 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 + B2 + C2 B + i C 1 3 3/2 + i 3 3 /2 1 3 E0 y e
This corresponds to elliptical polarization with
√
E0 x = A = 2, E0 y = B 2 + C 2 = 3 , ε = tan − 1 ( C/ B) = π/ 3
Then,
E0 x E0 y cos ε 2 ( 2) ( 3 ) cos ( π/ 3 )
tan 2 α = = ⇒ α = − 25. 07 ◦
E02 x − E02 y 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 3 e i π/ 3 1 /2 3 /2 3/2 + i 3 3 /2 ( 1 + 3 3 / 4) + i ( 9/ 4)
In general one can write,

a− ib a2 + b2
( 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,
" √ √ #
( 3 − 3√ / 4) − i ( 3 3 / 4) 2. 651 92 A
= =
( 1 + 3 3 / 4) + i ( 9/ 4) − 0. 66506 + i 5. 1 961 5 B+iC
√
So, E0 x = A = 2. 652, E0 y = B 2 + C 2 = 5. 239, ε = tan − 1 ( C/ B) = 97. 294 ◦ . Then
E0 x E0 y cos ε 2( 2. 65) ( 5. 24) cos ( 97. 3 ◦ )
tan 2 α = = ⇒ α = 4. 903 ◦
E02 x − E02 y ( 2. 65) 2 − 5. 24 2
The elliptical polarization is thus rotated by ( 4. 90 ◦ − ( − 25. 07 ◦ ) ) = +30 ◦ , as expected.
80
1 4-2 1 . According to Eq. ( 1 4 1 2) ,
2 2
Ex Ey Ex E y
+ −2 cos ε = sin 2 ε
E0 x E0 y E0 x E 0 y
( a) For ε = π/ 2 ⇒ cos ε = 0, sin ε = 1 . Then,

2 2
Ex Ey
+ =1
E0 x E0 y
This is an ellipse with semi axes of E0 x and E0 y aligned with the coordinate axes.
( b) If E02 x = E02 y = 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,
E0 x E0 y cos ε 2 E02 cos ε

tan 2α = = 2 = ∞ ⇒ 2 α = 90 ◦ ⇒ α = 45 ◦
E02x − E02 y 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
Ex E y
+ =0 −2
E0 x E0 y
E0 x E0 y
2
Ex E E0 y
− y = 0 ⇒ Ey = Ex
E0 x E0 y E0 x
This is an equation for a straight line of slope E0 y / E0 x .
1 4-2 2 . 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.

cos 2 α sin α cos α sin 2 θ − sin θ cos θ
=
sin α cos α sin 2 α − sin θ cos θ cos 2 α
Then the output light has the Jones vector,

sin 2 θ − sin θ cos θ 1 0 cos θ sin θ
= ( 1 − i) sin θ cos θ
− sin θ cos θ cos 2 α 0 i sin θ cos θ
A n aly z er QWP Amp= 1 Amp = 1
FA h or
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 sin 2 θ cos 2 θ | ( 1 − i) | 2 = 2 sin 2 θ cos 2 θ I0
81
C hapter 1 5 production of polarized light
1 5-1 . 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 cos 2 ( 30 ◦ ) = I0 cos 2 30 ◦
2
1
I3 = I2 cos 2 ( 60 ◦ − 30 ◦ ) = I0 cos 4 30 ◦ = 0. 281 5 I0 ⇒ 28. 1 5% of I0
2
n2
1 5-2 . The polarizing angle is given by the relation, tan θ p = . So for n a ir = 1 and n d iam = 2. 42
n1

n air 1
Internal reflection: θ p = tan −1
= tan −1
= 22. 5 ◦
n d iam 2. 42

n d iam 2. 42
External reflection: θ p = tan − 1 = tan − 1 = 67. 5 ◦
n air 1
1 5-3. ( a) See Figure 1 5 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 TA 1 be E0 p and let the amplitude of
the field component perpendicular to TA 1 be E0 n with E02 p + E02 n = E02 and E02 p = E02 n = E02 / 2. Then:
Polarizer 1 :
√ √
E1 p = α E0 p through TA 1 since the energy fraction is α. Similarly, E1 n = β E0 n through the extinction
axis EA 1 .
Polarizer 2: Now each component E1 p and E1 n has a component along both the transmission axis
TA 2 and extinction axis EA 2 of the second polarizer. Then
√ √
E2 p = α ( E1 p cos θ) thru TA 2 E2 n = α ( E1 n sin θ) thru TA 2
√ √ √ √
or E2 p = α ( α E0 p) cos θ thru TA 2 E2 n = 2 ( β E0 n ) sin θ thru TA 2
√ √
E2 p = β ( E1 p sin θ) thru EA 2 E2 n = β ( E1 n cos θ) thru EA 2
√ √ √ √
E2 p = β ( α E0 p) sin θ thru EA 2 E2 n = β ( β E0 n ) cos θ thru EA 2
Ip = ε 0 c/ 2) E p2 = ε 0 c/ 2) [ E2 p ( TA) ] 2 + [ E2 p ( EA) ] 2 In = ε 0 c/ 2) En2 = ε 0 c/ 2) [ E2 n ( TA) ] 2
2 2
Ip = ε 0 c/ 2) α E0 p cos θ + α β E02 P sin 2 θ In = ε 0 c/ 2) α β E02n sin 2 θ + β 2 E02n cos 2 θ
Ip = ε 0 c/ 2) E02 p ( α 2 cos 2 θ + α β sin 2 θ) In = ε 0 c/ 2) E02n ( α β sin 2 θ + β 2 cos 2 θ)
Using E02 p = E02n = E02 / 2 one finds,

ε c E2 1
I = Ip + In = 0 0
( α + β) cos 2 θ + 2α β sin 2 θ = I0 ( α + β) cos 2 θ + 2α β sin 2 θ
2 2 2
I = I0 ( 1 / 2) ( α + β) cos 2 θ + α β sin 2 θ
1
In the ideal case, β = 0 and α = 1 so that I = I0 cos 2 θ.
2
( b) With α = 0. 95, β = 0. 05, one finds

θ ideal actual 
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
1 5-4. = t ( ∆n) or t = = = 0. 063 mm
2 2 ∆n 2 ( 1 . 599 − 1 . 594)
1 5-5. See the answers in the back of the text.
1 5-6. ( a) With the flat plate in the position of maximum retardation,

2λ 2 ( 546. 1 × 1 0 − 6 mm
∆ m a x = 2 λ = ( ∆ n) ( d 2 − d 1 ) or d 2 − d 1 = = = 0. 1 2 mm
∆n 1 . 555 − 1 . 546
λ 546. 1 × 1 0 − 6 mm
( b) For circularly polarized light, ∆ = λ/ 4, so d 2 − d 1 = = = 0. 01 5 mm.
4 ∆n 4 ( 1 . 555 − 1 . 546)
1 5-7. ( a) The sketch is essentially the same as Figure 1 5 22 that accompanies problem 1 5 3 in the text. Then,
1
I1 = I0 / 2 E2 = E1 cos θ ⇒ I2 = I1 cos 2 θ = I0 cos 2 θ
2
1 1
( b) In this case, I2 = I0 cos 2 θ = I0 cos 2 ( 90 ◦ − 0 ◦ ) = 0.
2 2
( c) The angle between successive polarizers is always 1 5 ◦ so,
1 6
I7 = I0 cos 2 ( 1 5 ◦ ) = 0. 33 I0
2
So 33% of the incident light is transmitted.
1 5-8. The angular offset between successive polarizers is θ = 90 ◦ / N. Applying Malus’ law N times in succession,
N 2N
IT = I0 cos 2 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 .
1 5-9. Using, λ/ 4 = ( ∆n) t,

λ 589. 3 × 1 0 − 6 mm
t= = = 0. 01 62 mm
4 ∆ n 4 ( 1 . 5534 − 1 . 5443)
1 5-1 0. See Figure 1 5 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 ia gon al
0
θR
θR θ2
R ay
θ1
45 ◦
θ4
θ3
E xit fa c e
At the diagonal interface:

E p component from n k to n ⊥ : 1 . 4864 sin 45 = 1 . 6584 sin θ R or θ R = 39. 329 ◦
Es component from n ⊥ to n k : 1 . 6584 sin 45 = 1 . 4864 sin θ R0 or θ R0 = 52. 086 ◦
On exit:
Upper ray: θ 1 = 45 − θ R = 5. 671 ◦ ; 1 . 6584 sin 5. 671 ◦ = ( 1 ) sin θ 2 or θ 2 = 9. 432 ◦
Lower ray: θ 3 = θ R0 − 45 = 7. 086 ◦ ; 1 . 4864 sin 7. 086 ◦ = ( 1 ) sin θ 4 or θ 4 = 1 0. 566 ◦
Deviation: θ 2 + θ 4 = 9. 432 ◦ + 1 0. 566 ◦ = 1 9. 997 ◦ ≈ 20 ◦
83
1 5-1 1 . The arrangement is as shown below,
OA λ
∆ ϕ = π/ 2 or ∆ = = t ∆n
4
λ 600 × 1 0 − 6 mm
∆n = = = 0. 005
t 4t 4 ( 0. 03 mm)
1 5-1 2 . See Figure 1 5 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. 1 2 ◦
n1 1
( b) The angle θ R the refracted ray makes with the normal to the air/ water interface is

sin θ P
θ R = sin − 1 = 36. 877 ◦
1 . 333

0 1 . 50
The polarizing angle for the water/ glass interface is, θ P = 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
θ R = 36. 877 ◦ . However, for complete polarization off the glass, θ P0 must be 48. 37 ◦ . Thus the glass must be
tilted by 48. 37 ◦ − 36. 88 ◦ = 1 1 . 5 ◦ relative to the water surface.
1 5-1 3. ( 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.
1 5-1 4. ( a) The angle of incidence at which maximum polarization occurs is,

2. 42
θ P = tan − 1 = 56. 2 ◦
1 . 62
( b) The angle of refraction into the diamond is

1 . 62 sin θ P
θ R = sin −1
= 33. 8 ◦
2. 42
1 5-1 5. 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 / λ 2 .
( 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,
βr ed λ2
= v iolet ⇒ βv iolet = βred λ 2red / λ 2v iolet = ( 1 5 ◦ ) ( 700 2 / 400 2 ) = 46 ◦
βv iolet λ 2red
84
πz
1 5-1 6. In general the thickness z is related to rotation β by, β = ( nL − nR) .
λ0
β λ0 ( 0. 1 745) ( 396. 8 × 1 0 − 6 mm)
( a) If β = 1 0 ◦ = 0. 01 745 rad, z = = = 0. 200 m.
π ( ∆m) π ( 0. 0001 1 )
( b) Now, ρ = β for z = 1 mm. So,

πz π ( 1 mm)
ρ= ∆n = ( 0. 0001 1 ) = 0. 8709 rad = 49. 9 ◦
λ0 396. 8 × 1 0 − 6 mm
Also, by interoplation from Table 1 5 − 2, ρ ≈ 50 ◦ .
λ λ 589. 3 × 1 0 − 6 mm
1 5-1 7. ( a) = t ∆n ⇒ t = = = 8. 57 × 1 0 − 4 mm
4 4 ∆n 4 | 1 . 4864 − 1 . 6584|
( b) For Zircon, using Table 1 5 1 , n k = 1 . 968 and n ⊥ = 1 . 923. It is given that t = 0. 01 82 mm.
λ
To transmit, it must serve as a HWP so = t ∆n, or
z
λ = 2 t ∆ n = 2 ( 0. 001 82 cm) ( 1 . 968 − 1 . 923) = 1 . 638 × 1 0 − 4 cm
This is in the infrared. In general
t ∆n 8. 1 9 × 1 0 − 5 cm
( m + 1 / 2) λ = t ∆n ⇒ λ = =
m + 1 /2 m + 1 /2
8. 1 9 × 1 0 − 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.
1 5-1 8. Let,
n2 n
tan θ p = and tan θ p0 = 1
n1 n2
Then,
1
tan θ p = = cot θ p0
tan θ p0
So θ p = 90 ◦ − θ p0 . ( 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 = n 2 / n 1 − Brewster ’ s angle for light in.
θp
( 2) sin θ p = n sin θ R − Snell ’ s law.
θR Then from ( 1 ) : sin θ p = n cos θ p.

Substituting 2) : n sin θ R = n cos θ p: θ R and θ p must be complementary.
1 1 n
n = tan θ p = cot θ R = and tan θ R = = 1
tan θ R n n2
θp
85
1 5-1 9. ( a) ∆ = t ∆n. For t = λ , ∆ / λ = ∆ n = 1 . 5553 − 1 . 5462 = 0. 0091
λ 546 × 1 0 − 6 mm
( b) = t ∆ n or t = = 0. 01 5 mm = 1 5 µm
4 4 ( . 0091 )
t ∆n ( 0. 735 mm) ( 0. 0091 )
( c) m + 1 / 4) λ = t ∆ n ⇒ m + 1 / 4 = = = 1 2. 25 ⇒ m = 1 2
λ 546 × 1 0 − 6 mm
( d)
∆ 1 = − ∆ 2 ⇒ ∆ t ot al = t 1 ∆n + t 2 ( − ∆n) = λ/ 4
l 546 × 1 0 − 6 mm
t1 − t2 = = = 0. 01 5 mm = 1 5 µm
4 ∆n 4 ( 0. 0091 )
t1 t2
1 5-2 0. Given that λ = 500 nm and t = 1 / 1 6 in = 0. 0625 in = 0. 1 58875 cm,

( m + 1 / 2) λ ( m + 1 / 2) ( 500 × 1 0 − 6 mm)
( m + 1 / 2) λ ∆ n ⇒ ∆n = = = ( m + 1 / 2) ( 3. 1 496 × 1 0 − 4 )
t 1 . 5875 mm
So for ∆m = 1 between adjacent bands, ∆ n = 3. 1 5 × 1 0 − 4 .
λ0 ∆ ϕ
1 5-2 1 . From Eq. ( 1 5 3) , d = . From Table 1 5 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. 8 µ m
( b) Circular polarization: ∆ ϕ = and d = = 1 8. 4 µm
2 2
1 5-2 2 . From Table 1 5 2, ρ = 25. 535 deg/ mm at 546 nm. Then

β = ρ L = + 25. 535 deg/ mm × 1 . 1 5 mm = + 29. 365 ◦ ( ccw)
Thus the HWP must be set at β/ 2 = + 1 4. 68 ◦ , since it converts linearly polarized light at + α to to lin
early polarized light at − α.
1 5-2 3. The polarizing angle is θ p = tan − 1 n = tan − 1 n 2 / n 1 = tan − 1 1 . 5 = 56. 31 ◦ .
( a) The reflectance is " #2

√
cos 56. 31 − 1 . 5 2 − sin 2 56. 31
R= √ = 0. 1 479 ⇒ 1 4. 79%
cos 56. 31 + 1 . 5 2 − sin 2 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, 1 4. 8% is reflected, so 85. 2% is transmitted at each interface. The TM mode
is transmitted throughout at 1 00%. For N plates, there are 2 N interfaces, so for N = 1 0 we have
% TE transmitted = ( 0. 852) 2 N I0 / 2 = 0. 0203 I0 ⇒ 2. 03% I0
( c) Since IT M = 50% of I0 ⇒ IT M = 0. 50 I0 . So,

0. 5 − 0. 0203
P= = 0. 922
0. 5 + 0. 0203
1 5-2 4. See Figure 1 5 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 cos 2 ( 90 − 2 θ) ⇒ I = I0 sin 2 ( 2 θ) . his problem can
also be worked using matrices, in exactly the manner of problem 1 4 22, replacing the QWP matrix there by
the HWP matrix.
86
1 5-2 5. ( a) Using Eq. ( 1 5 1 0) and Table 1 5 3,
πz π ( 3 × 1 0− 3)
β= ( nL − nR) = β = ( 6 × 1 0 − 5 ) = 0. 742 rad = 42. 5 ◦
λ0 762 × 1 0 − 9
( b) From Eq ( 1 5 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 × 1 0− 5
β= ( nL − nR ) = = = 0. 01 05 rad = 0. 60 ◦
λ0 2| n ⊥ − n k | 2 | n k − n⊥ | 2 8. 94 × 1 0 − 3
87
C hapter 1 6 Holography
1 6-1 . See Figure 1 6 9 that accompanies the statement of this problem in the text and the sketch below. Using
Eq. ( 7 1 4) ,
√ √
I = I1 + I2 + I1 I2 cos δ = I = I1 + I2 + 2 I1 I2 2 cos 2 ( δ/ 2) − 1
with δ = 2 π ∆/ λ. From the sketch, ( s + ∆ ) 2 = r 2 + s 2 ⇒ s 2 + 2 s ∆ + ∆ 2 = r 2 + s 2 ⇒ ∆ ( 2s + ∆ ) = r 2

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

I1 √ π r2
s I = I1 + I2 + 2 I1 I2 2 cos 2
y=r 2λs
I2  
!
I2 √ √ π r2 
I = I1 + I2 − 2 I1 I2 +  4 I1 I2 cos 2
O
 2 λ s 
I2 B
I1 A α r2
I = A + B cos 2 ( αr 2 )
s
P late
√
1 6-2 . ( 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
1 6-3. If the storage could be done at the same density, the information stored holographically would be,
( 1 Gb) ( 1 mm/ 1 0 − 3 mm) = 1 000 Gb
λn 633 nm
1 6-4. d = = = 1 823 nm = 1 . 82 µm
2 sinθ 2 sin 1 0 ◦
x λ/ 1 0 633 × 1 0 − 1 0 m
1 6-5. v = = −9 = = 63. 3 m/ s
t 10 s 1 0− 9 s
2 2 2
Im ax AR + AS AR/ AS + 1 8+1
1 6-6. = = = = 1 . 65
Im in AR − AS AR/ AS − 1 8−1
V 1 mm 3 ( 1 0 − 3 ) 3 m3 10
1 6-7. Nb = 3 = ( 492 nm/ 1 . 30) 3 = ( 492 × 1 0 − 9 / 1 . 30) 3 m 3 = 1 . 8 × 1 0 bits
λn
1 6-8. ( a) Refer to Figure 1 6 5 in the body of the text and note that 2 θ = 1 80 ◦ . Then the spacing is given by
λn λ 500 nm
d= = 0 = = 1 56. 25 nm
2 2 n ( 2) ( 1 . 6)
( b) For m λ n = m λ 0 / n = 2 d sin 90 ◦ ⇒ . For m = 1 , λ 0 = 2 n d = 500 nm.
( c) In the film the angle with respect to the normal is sin ϕ F = ( 1 / 1 . 6) sin ( 30 ◦ ) ⇒ ϕ F = 1 8. 21 ◦
The Bragg angle is then θ = 90 ◦ − ϕ F = 71 . 79 ◦ . The Bragg condition then gives, ( for m = 1 ) ,
λ0
λn = = 2 d sin θ ⇒ λ 0 = 2 n d sin θ = 2( 1 . 6) ( 1 56. 25 nm) sin ( 71 . 79 ◦ ) = 475 nm
n
88
1 6-9. ( a) See the sketch in Figure 1 6 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 ϕ 0 = 30 ◦ . The angle with respect to the normal in the film is then
sin ϕ F = ( 1 / n) sin ϕ 0 = ( 1 / 1 . 6) sin ( 30 ◦ ) ⇒ ϕ F = 1 8. 21 ◦
so that the Bragg angle in the film is θ = 90 ◦ − ϕ F = 71 . 79 ◦ .
60 ◦ 60 ◦
1 8. 2 1 ◦ 30◦
30◦
71 . 79 ◦
( b) The Bragg condition for m = 1 gives,
λn λ0 633 nm
d= = = = 208 nm
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 ϕ F = 90 ◦ − 42. 54 ◦ = 47. 46 ◦ . The angle
of incidence from air with respect to the normal is given by Snell’ s law,
sin ϕ 0 = 1 . 6 sin ϕ F = 1 . 6 sin 47. 46 ◦ = 1 . 1 8
No solution is possible for this wavelength and this film.
λ 430
1 6-1 0. Assuming a film of index of refraction unity. The original spacing is d 0 = = = 21 5 nm. Upon
2 sin 90 2
shrinkage,
d = d 0 − 0. 1 5 d 0 = 1 82. 75 nm
Then, λ = 2 d sin 90 ◦ = 2 ( 1 82. 75 nm) ( 1 ) = 365 nm. The blue components shift into the ultraviolet and
appear missing.
λ R 633 λR 633
1 6-1 1 . ( a) m = = = 1 . 88 × ( b) m = = = 6330 ×
λ u v 337 λ x-r ay 0. 1 0
89
s i ( ωt + δ)
1 6-1 2 . ( a) The plane and spherical waves are of the form ER = r e i ω t and ES = e with EF = ER + ES .
R
Then,
s2 r s − iδ r s iδ
IF = | EF | 2 = ER2 + ES2 + ER ES + ER ES = r 2 + 2 + e + e
R R R
2 2 2
s r s r s i( ωt+ δ)
EH ∝ IF ER = r 2 + 2 ER + e i ( ωt − δ) + e
R R R
EH 1 EH 3 EH 2
Here EH1 represents the amplitude modulated reference beam, EH3 represents the phase reversed subject
beam ( real image) , and EH 2 represents the amplitude modulated subject beam ( virtual image) .
( b) The result δ = π y 2 / λd was derived in problem 1 3 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
10
1 ∆2 1 and 2 on the expanding wavefront for example
show lags related to the path difference:
2 20 ∆1 > ∆2 > ∆0 = 0 .
P C P0 When these differences are reversed, i. e. , promoted to

advances relative to C , then points such as 1 0 and 2 0 ,
shown along the direction of reinforcement, ( toward P 0 )
become appropriate. The points 0, 1 0 , 2 0 , . . . are consistent
with the spherical wavefront converging toward the
Plate ( real object) P 0 .
90
C hapter 1 7 O ptical Detectors and Displays
h c 1 240 eV · nm 1 240 eV · nm 1 240 eV · nm

1 7-1 . Egap = = ⇒λ= = = 1 850 nm
λ λ Eg ap 0. 67 eV
1 7-2 . ( a) 1 / 1 0= 0. 1 .
( b) Assume that N electrons are generated per second. Then the responsivity is
N e/ s e/ s 1 . 6 × 1 0 − 1 9 C/s
R= = 0. 1 = 0. 1 = 0. 0724 A/W
( 1 0 N h c/ λ) / s ( h c/ λ) / s ( 6. 626 × 1 0 − 3 4 · 3 × 1 0 8 / 0. 9 × 1 0 − 6 ) J/s

1 7-3. The responsivity is R = I/ P ⇒ I = R P = ( 0. 8 A/ W) 1 0 − 7 W = 0. 08 µA

Ep h ot on h c/ λ 6. 626 × 1 0 − 3 4 · 3 × 1 0 8 / 1 . 5 × 1 0 − 6
1 7-4. Pm in = = = W = 1 . 3 × 1 0− 1 3 W
detection time t d et 1 0− 6
1 7-5. ( a) The number of photons Np h in a detection time t d et is,

NEP 4 × 1 0− 1 1
Np h = = 3 4 = 300
( h c/ λ) / t d et ( 6. 626 × 1 0 − · 3 × 1 08/ 1 . 5 × 1 0 − 6) / ( 1 0− 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.
1 7-6. The transmission through the liquid crystal cell is dependent on the voltage across the cell as indicated in
Figure 1 7 9. Modulating the voltage then modulates the transmitted irradiance.
1 7-7. See for example, Saleh, B. E. A. , and M. C. Teich. Fundamentals of Photonics. New York: John Wiley and
Sons, 1 991 . Section 1 8 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 θ e .
Since the liquid crystal is birefringent, the index of refraction seen by light polarized along the θ = 0 direc
tion is
1 cos 2 θ e sin 2 θ e
= +
n2 ( θe ) n 2e n 2o
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.

1 7-8. The magnitude of the magnification is m = − s 0 / s = − 1 0 cm/ 1 00 cm = 0. 1 . Thus a feature of object
dimension 1 mm will have an image dimension of 0. 1 mm. 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.
1 7-9. The saturation power would be

Is at 1 0− 5 A
Ps at = = = = 0. 1 mW
R ( 0. 1 A/ W)
1 7-1 0. 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
C hapter 1 8 Matrix Methods in Paraxial O ptics
1 8-1 . See the sketch below. Using Eq. ( 1 8 1 ) ,

1 n − n 0 n L − n ( n L − n) ( n L − n 0 ) t
= L − −
f1 n R2 n R1 n nL R1 R2
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)
n0 1
f2 = − f1 = − ( − 62. 05 cm) = 46. 66 cm
n 1 . 33
Further, using Eq. ( 1 8 3) ,
nL − n 0 1.6 − 1
r= f1 t = ( − 62. 05 cm) ( 5 cm) = 2. 91 cm
nL R2 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 n0 = 1 . 33 n n= 1 F2
L
H2
f1
H1
f2
r s
1 8-2 . ( a) Using Eq. ( 1 8 1 )

1 n − n 0 n L − n ( n L − n) ( n L − n 0 ) t
= L − −
1 1 . 53 − 1 1 . 53 − 1 ( 1 . 53 − 1 ) 2 3 cm
= − − ⇒ f1 = 1 4. 06 cm
f1 1 2. 5 cm − 20 cm 1 . 53 ( 1 2. 5cm) ( − 20 cm)
n0
f2 = − f1 = − 1 ( − 1 4. 06 cm) = − 1 4. 06 cm
n
nL − n 0 1 . 53 − 1
r= f1 t = ( 1 4. 06 cm) ( 3 cm) = 1 . 1 7 cm
nL R2 1 . 53 ( 1 2. 5 cm)
n −n 1 . 53 − 1
s=− L f2 t = ( − 1 4. 06 cm) ( 3 cm) = − 0. 73 cm
nL R1 1 . 53( − 20cm)
( b) With “object distance” 30 cm from V1 or s = 31 . 1 7 cm from H1 ,

1 1 1 1 1 1
+ = ⇒ + = ⇒ s i = − 9. 689 cm from H2
s0 si f 31 . 1 7 cm s i − 1 4. 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 = − 1 4. 51 cm
f R1 R2 − 20 1 2. 5
1 1 1 1 1 1
+ = = + = ⇒ s i = − 9. 780 cm from lens center
s o s i f 30 cm s 0 − 1 4. 51 cm
The percent error in using the thin lens formula is
9. 78 − 8. 92
% error = = 9. 6%
8. 92
92
1 8-3. Using Eq. ( 1 8 1 )
1 n − n 0 n L − n ( n L − n) ( n L − n 0 ) t
= L − −
1 1.5 − 1 1.5 − 1 (1.5 − 1)2 5 cm
= − − ⇒ f1 = 1 2. 63 cm
f1 1 0 cm − 20 cm 1.5 ( 1 0 cm) ( − 20 cm)
n0
f2 = − f1 = − 1 ( 1 2. 63cm) = − 1 2. 63 cm
n
nL − n 0 1.5 − 1
r= f1 t = ( 1 2. 63 cm) ( 5 cm) = 2. 1 05 cm
nL R2 1 . 5 ( 1 0 cm)
n −n 1.5 − 1
s=− L f2 t = ( − 1 2. 63 cm) ( 5 cm) = − 1 . 0525 cm
nL R1 1 . 5( − 20cm)
Then s 0 = 8 cm + r = 1 0. 1 05 cm and,
f1 f2 1 2. 63 ( − 1 2. 63 cm)
− + =1⇒− + = 1 ⇒ s i = − 5. 61 4 cm left of H2
so si 1 0. 1 05 si
That is s i = − 5. 61 4 cm + s = − 6. 67 cm left of V2 . The magnification is,
n si ( 1 ) ( − 5. 61 4)
m=− =− = + 0. 556
n0 so ( 1 ) ( 1 0. 1 05)
So, h i = m h o = 0. 556 × 1 in = 0. 556 in.
1 8-4. Using Eq. ( 1 8 1 )

1 n − n 0 n L − n ( n L − n) ( n L − n 0 ) t
= L − −
1 1 . 61 − 1 . 33 1 . 61 − 1 ( 1 . 61 − 1 . 33) ( 1 . 61 − 1 ) 2 cm
= − − ⇒ f1 = − 1 1 . 51 cm
f1 ( 1 ) ( − 1 0 cm) ( 1 ) ( 1 0 cm) ( 1 ) ( 1 . 61 ) ( 1 0 cm) ( − 1 0 cm)
n0 1 . 33
f2 = − f1 = − ( − 1 1 . 51 cm) = 1 5. 31 cm
n 1
nL − n 0 1 . 61 − 1 . 33
r= f1 t = ( − 1 1 . 51 cm) ( 2 cm) = 0. 4 cm
nL R2 1 . 61 ( − 1 0 cm)
n −n 1.5 − 1
s=− L f2 t = ( 1 5. 31 cm) ( 2 cm) = − 1 . 1 6 cm
nL R1 1 . 61 ( 1 0cm)
Also, from Eq. 1 8 4,

n 0 nL − n 0 1 . 33 1 . 61 − 1 . 33
v= 1 − + t f1 = 1 − + ( 2) ( − 1 1 . 51 cm) = 4. 20 cm
n nL R2 1 1 . 61 ( − 1 0 )

n nL − n 1 1 . 61 − 1
w= 1 − 0− t f2 = 1 − − 2 ( 1 5. 31 cm) = 2. 64 cm
n nL R1 1 . 33 1 . 61 ( 1 0)
Now, s o = 60 cm + r = 60. 4 cm. So,
f1 f2 − 1 1 . 51 1 5. 31 cm
− + =1⇒− + = 1 ⇒ s i = 1 8. 9 cm from H2
so si 60. 4 si
The magnification is,
n si ( 1 ) ( 1 8. 9)
m=− =− = − 0. 235
n0 so ( 1 . 33) ( 60. 4)
So, h i = m h o = ( − 0. 235) ( 5 cm) = − 1 . 1 8 cm.
93
1 8-5. ( a) Using Eq. ( 1 8 1 )
1 n − n 0 n L − n ( n L − n) ( n L − n 0 ) t
= L − −
1 1 . 33 − 1 1 . 33 − 1 ( 1 . 33 − 1 ) 2 20 cm
= − − ⇒ f1 = − 20. 1 5 cm
f1 ( 1 ) ( − 1 0 cm) ( 1 ) ( 1 0 cm) ( 1 ) ( 1 . 33) ( 1 0 cm) ( − 1 0 cm)
n0 1
f2 = − f1 = − ( − 20. 1 5 cm) = 20. 1 5 cm
n 1
Using Eq. ( 1 8 3) ,
nL − n 0 1 . 33 − 1
r= f1 t = ( − 20. 1 5 cm) ( 20 cm) = 1 0 cm
nL R2 1 . 33( − 1 0 cm)
n −n 1 . 33 − 1
s=− L f2 t = − ( 20. 1 5 cm) ( 20 cm) = − 1 0 cm
nL R1 1 . 33( 1 0cm)
Since n = n 0 = 1 , v = r = 1 0 cm and w = s = − 1 0 cm.
( b) The object distance is s o = 20 cm + 1 0 cm = 30 cm from H1 . So,

f1 f2 − 20. 1 5 20. 1 5 cm
− + =1⇒− + = 1 ⇒ s i = 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
ob ject
F2
F1
image
H2 , N2
1 8-6. ( a) The system matrix is

     
1 0 1 5 1 0 1 1 0/ 3
M = R s p h TR p lan e =     = 
− 1 /1 0 3/2 0 1 0 2/ 3 − 1 /1 0 2/ 3
y = A y0 + B α 0 = ( 1 ) ( 1 cm) + ( 1 0/ 3 ) ( 0) = 1 cm
α = C y0 + D α 0 = ( − 1 / 1 0) ( 1 cm) + ( 2/ 3 ) ( 0) = − 0. 1 rad = − 5. 73 ◦
With an additional translation,
    
1 x 1 1 0/ 3 1 − x/ 1 0 1 0/ 3 + 2 x/ 3
M 0 = T M =   = 
0 1 − 1 /1 0 2/ 3 − 1 /1 0 2/ 3
y = ( 1 − x/ 1 0) y0 + ( 1 0/ 3 + 2 x/ 3 ) α 0
α = − 0. 1 y0 + 2 α 0 / 3
( c) The ray crosses the axis when y = 0. For y0 = 1 , α 0 = 0, y = 0 ⇒ 1 − x/ 1 0 = 0 ⇒ x = 1 0 cm
94
1 8-7. Use y f = A y0 + B α 0 and α f = C y0 + D α 0 . From Figure 1 8 1 2b, a 0 = 0, y0 = y0 , y f = y f so,
y f = A y0 , α f = C y0
Proceeding, y f = A y0 + B α 0 , α f = C y0 + D α 0
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 1 8 1 2c, α 0 = 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 − n 0 / 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
1 8-8. 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 / 1 3 25/ 26
1 . 3 ( 1 . 5) 1.3 1 . 6 ( 1 . 5) 1.6
(1) (2)
p = D/ C = − 25/ 6 = − 4. 1 7 cm q = − A/ C = 1 3/ 6 = 2. 1 7 cm
n= 1.6
D − 1 /1 .3
n= 1 n = 1 . 33
r= = − 0. 83 cm s = ( 1 − A) / C = − 2. 1 7 cm
2 cm C
n0 / n f 1
f1 = = − 3. 33 cm f2 = − = 4. 33 cm
C C
R 1 = 1 . 5 cm R 2 = 1 . 5 cm
1 8-9. The lens matrix is,

     
1 0 1 5 1 0 1 /2 5/ 4
    = 
1 /1 0 1 0 1 − 1 /1 0 1 − 1 / 20 3/2
f1 = 1 / C = − 20 cm, f2 = − 1 / C = 20 cm
q = − A/ C = 1 0 cm, p = D/ C = − 30 cm
r = ( D − 1 ) / C = − 1 0 cm, s = ( 1 − A) / C = − 1 0 cm
1 8-1 0. The lens matrix is,

     
1 0 1 3 1 0 4/ 5 2
     
     
 0. 1 1 . 5   − 0. 5 1  = 
0 1 − 3 / 70 1 1 /14
1 . 4 ( 2) 1.4 1 . 5 ( 5) 1.5
f1 = ( n 0 / n f ) / C = ( 1 / 1 . 4) / C = − 1 6. 67 cm, f2 = − 1 / C = 25. 33 cm
q = − A/ C = 1 8. 67 cm, p = D/ C = − 1 8. 33 cm
r = ( D − n 0 / n f ) / C = − 1 . 67 cm, s = ( 1 − A) / C = − 4. 67 cm
95
1 8-1 1 . ( a) , ( b) The lens matrix is,
     
1 15 1 0 1 30 − 1 /2 0
    = 
0 1 − 1 /1 0 1 0 1 − 1 /1 0 −2
f1 = 1 / C = − 1 0 cm, f2 = − 1 / C = 1 0 cm
q = − A/ C = − 5 cm, p = D/ C = 20 cm
r = ( D − 1 ) / C = 30 cm, s = ( 1 − A) / C = − 1 5 cm
v = ( D − 1 ) / C = 30 cm, w = ( 1 − A) / C = − 1 5 cm
The principal and nodal points fall at the lens center and the focal points fall 1 0 cm to either side of the
lens.
( c) This is case ( c) of Figure 1 8 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
+ = ⇒ s i = 1 5 cm
30 cm s i 1 0 cm
Then m = − s i / s o = − 1 5/ 30 = − 1 / 2 = A.
1 8-1 2 . ( a) The lens matrix is,

     
1 15 1 8 1 30 1 /3 1 6/ 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.
1 8-1 3. ( a) , ( b) The lens matrix is,

     
1 0 1 1 1 0 1 6/ 1 5 2/ 3
     
    = 
   − 0. 5   
0. 5/ ( − 5) 3/2 0 1 2/ 3 − 1 / 1 50 1 4/ 1 5
( 1 . 5) ( − 5 )
f1 = 1 / C = − 1 50 cm, f2 = − 1 / C = 1 50 cm
q = − A/ C = 1 60 cm, p = D/ C = − 1 40 cm
r = ( D − 1 ) / C = 1 0 cm, s = ( 1 − A) / C = 1 0 cm
96
1 8-1 4. ( a) , ( b) The lens matrix is
       
 1 0   1 1 /2 1 0 1 1 1 0 0. 9764 0. 96755
       
     
 0. 62   − 0. 1 1 . 52   − 0. 52 1  = 
1 . 62 0 1 0 1 0. 0091 82 1 . 0333
20 − 20 ( 1 . 62) 1 . 62 ( 20) ( 1 . 52) 1 . 52
f1 = 1 / C = − 1 50 cm, f2 = − 1 / C = 1 50 cm
q = − A/ C = 1 60 cm, p = D/ C = − 1 40 cm
r = ( D − 1 ) / C = 1 0 cm, s = ( 1 − A) / C = 1 0 cm
( c) Consider each lens separately in air:

1 n2 − n1 1 1 1 1 . 52 − 1 1 1
= − ⇒ = − = 0. 052 cm − 1
1 . 62 f n1 R1 R2 f1 1 20 cm − 20 cm

1 1 . 62 − 1 1 1
1 . 52 = − = − 0. 062 cm − 1
f2 1 − 20 cm 20 cm
1 1 1
Equivalent focal length: = + = 0. 052 cm − 1 − 0. 062 cm − 1 ⇒ feq = − 1 00 cm
feq f1 f2
1 8-1 5. ( a) Using the matrix calculated in the text,

     
1 s0 2/ 3 2 1 s 2/ 3 − s 0 / 6 2/ 3 s + 2 − s s 0 / 6 + s 0
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,
4s + 12
2/ 3 s + 2 − s s 0 / 6 + s 0 = 0 ⇒ s 0 =
s−6
When B = 0, A = m = 2/ 3 − s 0 / 6.
( c) When A = 2/ 3 − s 0 / 6 = 0, the output plane at s 0 = 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.
1 8-1 6. 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 − n 3 1   n2 − 1  
n3 0 1 0 1 n2 0 1
r6 n 3 r5 n3 r4
       
1 0 1 d1 1 0 1 t1 1 0 A B
       
×  1 − n 
1 

 n 1 − 1



 1 − n 1
 
1  =


2
0 1 n1 0 1 C D
n 2 r3 n2 r2 n 1 r1 n1
Matrix multiplication and insertion of parameter values gives,
A = 0. 93935 , B = 22. 221 2 , C = 0. 009284, D = 0. 8448
and
D A D−1 1−A
p= = − 90. 99 mm, q = − = 1 01 . 1 8 mm, r = v = = 1 6. 72 mm, s = w = = − 6. 53 mm
C C C C
f1 = 1 / C = − 1 07. 71 mm, f2 = − 1 / C = 1 07. 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 8-1 7. The lens matrix is,
      
1 0 1 t 1 0 1 d2 A B
      
     = 
 n − n0 n L   n − n L n    
L
0 1 0 1 C D
n 0 R2 n0 nL R1 nL
Performing the matrix multiplication gives,

n − nL nt n nL − n 0 n t
A= t+ 1, B = , D= 0+
nL R1 nL n n 0 R2 nL
n L − n 0 n L − n ( n L − n) ( n L − n 0 ) t
C= − 0 −
n 0 R2 n R1 n 0 nL R1 R2
n 1 1 n0
1 8-1 8. From Table 1 8 2, f1 = 0
⇒ = C. Substituting C from problem 1 8 1 7,
n C f1 n

1 n 0 n L − n 0 n L − n ( n L − n) ( n L − n 0 ) t
= − 0 −
f1 n n 0 R2 n R1 n 0 nL R1 R2
1 n − n 0 n L − n ( n L − n) ( n L − n 0 ) t
= L − −
1 8-1 9. From Table 1 8 2 and using the results of problem 1 8 1 7,

h i
n nL − n 0 nt n
D − n/ n 0
n0
+ n 0 R2 nL
− n0
r= =
C C
n 1
but, f1 = . So,
n0 C

nL − n 0 nt n0 nL − n 0
r= f1 = f1 t
n 0 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 R1 nL R1
and,

n nL − n 0 nt 0
D−1 n0
+ n 0 R2 nL
−1 n nL − n 0 n t n n 0 nL − n 0
v= = = + − 1 f1 = 1 − + t f1
C C n0 n 0 R2 nL n n nL R2

( n/ n 0 ) − A n n − nL n nL − n
w= = − t − 1 ( − f2 ) = 1 − 0 − t f2
C n0 nL R1 n nL R1
98
1 8-2 0. A typical program in “Basic” which can be easily adapted to other languages:
1 0 REM: MERI DI O NAL RAY TRACI NG. AFTER THE LAST SURFACE I S CALCULATED, ENTER ANY
20 REM: NEGATI VE NUMBER AS DI STANCE TO NEXT SURFACE TO TERMI NATE
30 I NPUT 00 LEFT AND RI GHT REFERENCE I NDI CES: 00 , N1 , N2
40 I NPUT 00 RAY ANGLE DEGREE AND HEI GHT: 00 , A, H
50 I NPUT 00 DI STANCE FRO M O BJECT PO I NT ( I F I NFI NI TY ENTER ANYTHI NG) : 00 , D
60 REM: FO R RAY ENTERI NG PARALLEL TO AXI S, D I S NOT USED I N THE CALCULATI O N
70 P=3. 1 41 5926#: A=P* ( A/1 80) : REM ANGLE I S NOW I N RADI ANS
80 I F A=0 THEN GO TO 1 1 0
90 S=D H/ TAN( A) : Q = S* SI N( A)
1 00 GO TO 1 40
1 1 0 Q =H
1 20 REM: FO R PLANE SURFACE, ENTER R> 99990.
1 30 REM: THE ACTUAL VALUE I S NOT USED I N THE CALCULATI O N.
1 40 I NPUT 00 I NPUT RAD O F CURV ( 99999 I F PLANE) : 00 , R
1 50 I F R> 99990! THEN GO TO 21 0
1 60 B=Q /R+SI N( A) : T=ATN( B/ SQ R( 1 B* B) ) : REM TAKI NG ARCSI N( B)
1 70 C=N1 * SI N( T) /N2: T1 =ATN( C/ SQ R( 1 C* C) ) : REM TAKI NG ARCSI N( C)
1 80 A1 =T1 T+A: A2=( A1 / P) * 1 80: REMA2 CO NVERTS A1 TO DEGREES AGAI N
1 90 Q 1 =( SI N( T1 ) SI N( A1 ) ) * R: S1 = Q 1 / SI N( A1 )
200 GO TO 240
21 0 X=N1 * SI N( A) /N2: A1 =ATN( X/ SQ R( 1 X* X) ) : REM TAKI NG ARCSI N( X)
220 Q 1 =Q* CO S( A1 ) / CO S( A)
230 S1 = Q 1 / SI N( A1 ) : A2=( A1 / P) * 1 80: REM A2 I S A1 I N DEGREES
240 PRI NT 00 RAY I NTERSECTS AXI S AT 00 ; S1 ; 00 FRO M SURFACE. 00
250 PRI NT 00 RAY ANGLE I N RADI ANS I S 00 ; A1
260 PRI NT 00 RAY ANGLE I N DEGREES I S 00 ; A2
270 PRI NT 00 THE Q PARAMETER I S 00 ; Q 1
280 A=A1 : N1 =N2
290 I NPUT 00 DI STANCE TO NEXT SURFACE: 00 , D1
300 I F D1 < 0 THEN 360
31 0 Q =Q 1 +D1 * SI N( A1 )
320 I NPUT 00 NEW REFRACTI VE I NDEX: 00 , N2
330 I NPUT 00 NEW RADI US OF CURV ( 99999 I F PLANE) : 00 , R
340 I F R> 99990! THEN GO TO 21 0
350 GO TO 360
360 END
1 8-2 1 . Use the computer program ( problem 1 8 20) or Table 1 8 3 in a two step process for each ray, yielding
α = 0 ◦ : s 0 = 3. 1 80 cm and α 0 = − 23. 51 ◦
α = − 20 ◦ : s 0 = 1 6. 1 04 cm and α 0 = 6. 081 ◦
1 8-2 2 . Use the computer program ( problem 1 8 20) or Table 1 8 3 in a 3 step process, giving,
s 0 = − 49. 525 cm and α 0 = 3. 371 ◦
1 8-2 3. Use the computer program ( problem 1 8 20) or Table 1 8 3 in a 6 step process, to get,
h = 1 : s 0 = 98. 20 mm and α 0 = − 0. 567 ◦
h = 5: s 0 = 1 02. 45 mm and α 0 = − 2. 723 ◦
99
C hapter 1 9 O ptics of the Eye
1 9-1 . ( a) The conversion between the radiant flux and the luminous flux is
Φ v = K( λ) Φ e = 685 ( λ) Φ e
Estimating from Figure 1 9 2,
Φ v , H e-C d 685 ( 0. 02) ( 0. 050)
= = 1 . 25
Φ v , H e-N e 685 ( 0. 2) ( 0. 004)
( b) For equal brightnesses,

Φ v , A r = Φ v , H e- N e ⇒ 685 V( λ A r ) Φ e , A r = 685 V( λ H e- N e Φ e , H e-N e
( 0. 2) Φ e = ( 0. 95) ( 0. 5 mW)
Φ e , A r = 2. 4 mW
1 9-2 . ( a) From Table 1 9 1 ,

Iv dΦ v / dω Φ v/ 4 π
= = ⇒ Iv = Ev r 2 = ( 1 00 lm/ m 2 ) ( 3 m) 2 = 900 cd
Ev dΦ v / dA Φ v / ( 4 π r 2 )
( b) See Eq. ( 1 1 3) , Iv ( θ) = Iv ( 0) cos θ.
S ou rc e Using the general result in ( a) , at point Q In the diagram,

√ √
r= 10
3m I ( 0) cos θ ( 900 lm) ( 3 / 1 0 )
θ Ev = v 2 = √ 2 = 85. 4 lx
r
P Q F lo or 10 m
1m
1 9-3. Use the relation between illuminance and luminous intensity given in the solution to problem 1 9 2a:
Iv ( 0) cos ( 0) Iv ( 0) 3 / 5
Ev ( P) = + = 1 . 351 × 1 0 − 3 / ft 2 Iv ( 0)
√ ( 30 ft) 2 ( 50 ft) 2
S1 1 300 0 S2  √ 
50 0
 v I ( 0) ( 30/ 1 300 
30 0 Ev ( Q) = 2 √ 2  = 1 . 280 × 1 0 − 3 / ft 2 Iv ( 0
1 300 ft
Ev ( P)
P
20 0 Q = 1 . 055
Ev ( Q )
1 9-4. The arrangement is sketched ( not to scale) below,
The indicate angle and solid angle satisfy the relations

50 cm cos θ 0 = cos( sin − 1 ( 5 / 50) = 0. 9950, Ω 0 = 2 π ( 1 − cos θ 0 = 0. 01 π sr
θ0 5 cm
r
A F Ω0 Φ v , in c = Ω 0 Iv = ( 0. 01 π) ( 1 00) = π lm, Φ v , ref = 0. 8 Φ v , in c = 0. 8 π lm

Φ v , ref 0. 8 π lm
Ev = = = 320 lx
A π ( 0. 05 m) 2
1 00
1 9-5. ( a) The solid angle intercepted is Ω = 2 π ( 1 − cos θ) = 2 π ( 1 − cos( 0. 25 ◦ ) ) = 5. 98 × 1 0 − 5 sr. Here θ is the half
angle intercepted by the sun viewed from the earth. The luminance of the sun is about,
Ev 1 0 5 lm/ m 2
Lv = = = 1 . 67 × 1 0 9 cd/ m 2
Ω 5. 98 × 1 0 − 5 sr
θ
E S

lm
( b) Ev = L v Ω = L × 2 π sr = 2 π L lm/ m 2 .
m 2 sr
1 9-6. The arrangement is sketched below.

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

Z Z Z
L cos θ 1 cos θ 2 cos θ 1 cos θ 2
∆Φ1 2 = 2 dA 1 dA 2 = L ∆A 2 dA 1
A1 A2 r 12 A1 r12 2
Consider first illumination of ∆A 2 by an annular ring of incremental area dA 1 = 2 π r dr, as shown. Then,
√
r 1 2 = r 2 + D 2 , cos θ 1 = D/ r 2 + D 2
√
2 D r
cos θ 2 = cos( 45 − θ 1 ) = cos( 45 ) cos θ 1 + sin ( 45 ) sin θ 1 =
◦ ◦ ◦ √ +√
2 r2 + D 2 r2 + D 2
Adding the contributions from the annular rings that make up the disc,
ZR
cos θ 1 cos( 45 ◦ − θ 1 )
∆Φ 1 2 = L ∆A 2 2 π r dr
0 r2 + D 2
ZR √
D 2 D r 2 πr
∆ Φ 1 2 = L ∆ A2 √ √ +√ dr
0 r +D 2
2 2 2
r +D 2 2
r +D 2 r + D2
2
( ZR ZR )
√ r dr r 2 dr
∆Φ1 2 = 2 π L D ∆A 2 D +
0 ( r2 + D 2 ) 2 0 ( r + D2) 2
2
( R R)
√ −1 −r 1 r
∆Φ 1 2 = 2 π L D ∆A 2 D + + tan − 1
2 ( r2 + D 2 ) 0
2 2
2 (r + D ) 2 D D 0

√ 1 −1 R 1 D+R
∆Φ1 2 = 2 π L D ∆ A2 tan + −
2D D 2 D 2( R 2 + D 2 )
With L = 1 0 5 lm/ sr m 2 , D = 1 m, R = 0. 2 m, dA 2 = 1 0 − 4 m 2 ,
∆Φ 1 2 = 0. 97 lm
n2 − n1
1 9-7. The power for a refracting surface is defined as P = . So,
R
nA . H − 1 1 . 33 − 1
P= = = 41 . 6 m
R 8 × 1 0− 3 m
This value is in agreement with Table 1 9 2.
1 01
1 9-8. From Table 1 9 2: n len s = 1 . 45, R 1 = 1 0 mm, and R 2 = − 6 mm.
( a) Using the lensmaker’ s formula,

1 n len s − 1 1 1 1 . 45 − 1 1 1 1 1
= − = − ⇒ f = 8. 33 mm = P = = = 1 20 D
f 1 R1 R2 1 1 0 mm − 6 mm fin m 0. 00833 m
( b) In its environment,

1 n len s − n A H 1 1 1 . 45 − 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 : s 20 = ∞ , s 1 = f1 ( measured from the first surface) . Then,

R1 = 1 0 m m
R2 = − 6 m m n len s n A H n A H − n len s 1 1 . 33 − 1 . 45
+ = ⇒ = = s 2 = 66. 7 mm
nA H = 1 . 3 3 nA H = 1 . 3 3 s2 ∞ R2 s 2 ( 1 . 33) ( − 6 mm)
F2
s 10 = ( − 66. 7 − 3. 6) mm = − 63. 1 mm.
F1 f2 f1
nL
n A H n len s n len s − n A H
nle n s = 1 . 45 + 0 = ⇒ s 1 = f1 = 38 mm
3.6 mm s1 s1 R1
Proceeding in the same way for f2 :
s 1 = ∞ , s 20 = f2 ( measured from the second surface)
n A H n len s n len s − n A H
+ 0 = ⇒ s 10 = 1 20. 8 mm
∞ s1 R1
s 2 = − ( 1 20. 8 − 3. 6) mm = − 1 1 7. 2 mm
n len s n A H n len s − n A H
+ = ⇒ s 2 = 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 1 8, The lens matrix is,
     
1 0 1 3. 6 1 0 0. 9702 3. 3021
     
     
 1 . 45 − 1 . 33 1 . 45   1 . 33 − 1 . 45 1 . 33  = 
0 1 − 0. 02361 0. 9503
( 1 . 33) ( − 6 ) 1 . 33 ( 1 . 45) ( 1 0) 1 . 45
S u rfac e R 2 Tran s la t ion S u rfac e R 2
q = − A/ C = 41 . 09 mm, p = D/ C = − 40. 25 mm
r = ( D − 1 ) / C = 2. 1 05 mm, s = ( 1 − A) / C = − 1 . 262 mm
f1 = 1 / C = − 42. 355 mm f2 = − 1 / C = 42. 355 mm
1 02
1 9-9. Refer to Figure 1 9 3 and Table 1 9 2.
( a) For an unaccommodated eye the object is at infinity. Then at the cornea

n1 n2 n2 − n1 n2 R ( 1 . 33) ( 8 mm)
+ 0 = ⇒ s0 = = = 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.
n A H n len s n len s − n A H 1 . 33 1 . 45 1 . 45 − 1 . 33
+ = = + 0 = ⇒ s 0 = 24. 73 mm ( right of 1 st lens surface)
s s0 R1 − 28. 4 mm s 1 0 mm
At the rear surface of the lens, s = ( 3. 6 − 24. 73 ) mm = − 21 . 1 3 cm:
n len s n A H n A H − n len s 1 . 45 1 . 33 1 . 33 − 1 . 45
+ 0 = ⇒ + 0 = ⇒ s 0 = 1 5. 1 3 mm ( right of 2 n d lens surface)
s s R2 − 21 . 1 3 mm s − 6 mm
Thus an object at infinity forms an image at 1 5. 1 3 mm + 7. 2 mm = 22. 33 mm from the cornea the back focal
plane of the eye. This value is in agreement with Table 1 9 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
+ 0 = ⇒ s 0 = 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
+ 0 = ⇒ s 0 = 23. 83 mm ( right of front lens surface)
− 32. 53 mm s 6 mm
At the rear lens surface, s = ( 4 − 23. 83) mm = − − 1 9. 83 mm:
1 . 45 1 . 33 1 . 33 − 1 . 45
+ 0 = ⇒ s 0 = 1 4. 39 mm ( right of back lens surface)
− 1 9. 83 s − 6 mm
So, for an eye accommodated for the normal near point, the image is formed at
( 1 4. 39 + 7. 2) mm = 21 . 59 mm form the cornea.
1 9-1 0. ( 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) ( 1 0) 1 . 45 ( 1 . 33) ( 8 ) 1 . 33
R ea r len s su rfac e L en s Tran s. Front len s su rfac e Tran s. t o c orn ea C orn ea s u rfac e
 
0. 75846 5. 1 050
M = 
− 0. 0501 2 0. 651 80
q = − A/ C = 1 5. 1 33 mm, p = D/ C = − 1 3. 005 mm
r = ( D − 1 ) / C = − 1 . 963 mm, s = ( 1 − A) / C = − 4. 820 mm
f1 = ( n 0 / n f ) / C = ( 1 / 1 . 33) / C = − 1 5. 00 mm f2 = − 1 / C = 1 9. 95 mm
The first principal plane ( H in Figure 1 9 3) is a distance | r | = 1 . 96 mm to the right of the front corneal sur
face. The second principal plane ( H 0 in Figure 1 9 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 1 9 3. The front focal point is a distance of 1 5. 0
mm to the left of the H which places it a distance of 1 3. 0 mm from the front corneal surface. The back
focal plane is a distance of 1 9. 95 mm to the right of H 0 which places to the right of the corneal surface by a
distance of 1 9. 95 mm + 2. 38 mm = 22. 3 mm. These positions also agree with the values given in Table 1 9 3.
1 03
1 9-1 1 . In a Snellen chart letters in the 20/ 20 row should subtend an angle of θ l = 5 0 = 5 / 60 ◦ = 0. 001 454 rad and
details should subtend an angle θ d = 1 0 = 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 θ l = ( 60 in) ( 0. 001 454 rad) = 0. 0872 in, yd = yl / 5 = 0. 01 74 in.
20/1 5 row: yl = ( 1 5/ 20) yl , 2 0 / 2 0 = 0. 00654 in, yd = yl / 5 = 0. 001 31 in.
20/300 row: yl = ( 300/ 20) yl , 2 0 / 2 0 = 1 . 308 in, yd = yl / 5 = 0. 262 in.
20/1 00 row: yl = ( 1 00/ 20) yl , 2 0 / 2 0 = 0. 436 in, yd = yl / 5 = 0. 0872 in.
20/60 row: yl = ( 60/ 20) yl , 2 0 / 2 0 = 0. 262 in, yd = yl / 5 = 0. 0523 in.
1 9-1 2 . Objects 25 cm from the eye should form images 1 25 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 0 = − 1 23. 5 cm.
( a) The needed power is then
1 1 1 1 1
P= = + = − = 3. 45/ m = 3. 45 D
f s s 0 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 s far from the glasses is
found as,
1 1 1
P= = + ⇒ s fa r = f = 1 / P = ( 1 / 3. 45) m = 0. 290 m = 29. 0 cm
f s far − ∞
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.
1 9-1 3. ( 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 0 ∞ s 0 − 0. 5 m
( b) The new near point is the object position that places an image 1 5 cm from the eye:
1 1 1 1
P = − 2/ m = + = − ⇒ s = 0. 21 4 m = 21 . 4 cm
s s 0 s 0. 1 5 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 0 ∞ s 0 − 0. 48 m
1 1 1 1
P = − 2. 083/ m = + 0 = − ⇒ s = 0. 1 78 m = 1 7. 8 cm
s s s 0. 1 3 m
The needed power is − 2. 083 D and the near point is 1 7. 8 cm from the glasses or 1 9. 8 cm from the eye.
1 9-1 4. ( 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
1 04
1 9-1 5. 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 = + 0 = + ⇒ s0= f = = = − m = − 1 4. 3 cm
f s s ∞ s0 P − 7D 7
The unaided far point is 1 4. 3 cm from the eye.
1 1 1 1 1
Right eye near point: P = = − 7D= + 0 = + ⇒ s 0 = − 0. 0732 m = − 7. 32 cm
f s s 0. 1 5 m s 0
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 = + 0 = + ⇒ s0= f = = = − m = − 20 cm
f s s ∞ s0 P −5D 5
The unaided far point is 20 cm from the eye.
1 1 1 1 1
Left eye near point: P = = − 5 D= + 0 = + ⇒ s 0 = − 0. 0857 m = − 8. 57 cm
f s s 0. 1 5 m s 0
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= + 0 = + ⇒ s = 0. 502 m = 50. 2 cm
f s s s − 0. 1 43 m
The far point with the wrong lens is 50. 2 cm.
1 1 1 1 1
Right eye near point: P = = − 5D= + 0 = + ⇒ s = 0. 1 1 5 m = 1 1 . 5 cm
f s s s − 0. 0732 m
The near point with the wrong lens is 1 1 . 5 cm.
1 1 1 1 1
Left eye far point: P = = − 7D= + 0 = + ⇒ 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= + 0 = + ⇒ s = 0. 21 4m = 21 . 4 cm
f s s s − 0. 0857 m
The near point with the wrong lens is 21 . 4 cm.
1 9-1 6. 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 ( 1 5 − 1 . 5) cm from the lens:
1 1 1 1
Pfar = + = + = − 7. 41 D
s s 0 ∞ − 0. 1 3. 5 m
With this lens alone the corrected near point would be the object distance leading to images 1 1 . 5 cm from
the lens:
1 1 1
= P − 0 = − 7. 41 D − ⇒ s = 0. 778 m
s s − 0. 1 1 5 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
Pn ea r = + = + = − 6. 1 0 D
s s 0 0. 385 m − 0. 1 1 5 m
1 05
C hapter 2 0 Aberration Theory
20-1 . Using Eq. ( 20 6) ,

a( Q ) = ( n 1 l + n 2 l 0 ) − ( n 1 s + n 2 s 0 ) (1)
Eqs. ( 20 1 6) and ( 20 1 7) are,
" #
h 2 ( R + s ) h 4( R + s ) h 4 R + s ) 2
l=s 1+ + − ( 2)
2 R s2 8 R3 s 2 8 R2 s 4
" #
0 0 4 0 2
h 2
( R − s ) h 4
( R − s ) h R − s )
l0 = s0 1 + + − 0 (3)
2 R s 02 8 R3 s 02 8 R2 s 4
Using these in ( 1 ) and doing some obvious simplification gives,
" # " #
0 2
h 2 ( R + s ) h 4( R + s ) h 4 R + s ) 2 h 2
( R − s 0
) h 4
( R − s 0
) h 4
R − s )
a( Q ) = n 1 s + − + n2 s 0 + − 0 ( 4)
2 R s2 8 R3 s 2 8 R2 s 4 2 R s 02 8 R3 s 02 8 R2 s 4
Consider the terms proportional to h 2 :

h 2 n1 ( R + s ) n2 ( R − s 0) h 2 n1 n2 n1 − n2
+ = + 0 + =0
2 Rs Rs0 2 s s R
Now the h 4 terms:

h4 n1 n1 n1 n1 n1 n2 n2 n2 n2 n2
+ − − 2 − + − − + 2 − =
8 R2 s R3 s 3 R s 2 R2 s R2 s 0 R3 s 03 R s 0 2 R 2)s 0
( 2 2
h4 n1 1 1 n1 1 1 1 n1 n2 n1 n2
− + − 0 − + + 0 + −
8 s s R s s0 R R2 s s R R
( )
2 2
h 4 n1 1 1 n 1 1
a( Q ) − + + 10 −
8 s s R s s0 R
20-2 . ( a) If
n1 n2 n2 − n1 1 1 1
+ 0 = and 0 = +
s s R s s R
then,
n1 n2 1 1

n n n n
+ 0 = ( n2 − n1 − = 20 − 2 − 10 + 1
s s s0 s s s s s
n2 n1 0
= − 0 ⇒ s = − ( n1 / n2 ) s
s s
( b) In a( Q ) the h 2 term is zero due to Gaussian imaging. For the remainder we must examine
( 2 2)
h 4 n1 1 1 n2 1 1
a( Q ) = − + + 0 −
8 s s R s s0 R
Substitute using 1 / s 0 = 1 / s + 1 / R, to give,
( 2 2)
h 4 n1 1 1 1 n2 1 1 1 h 4 n1 n2 n1 s + n2 s 0
a( Q ) = − − + + 0 + − =− + = =0
8 s s0 R R s s R R 8 s s 02 s 0s 2 s 02 s 2
( c) For s 0 = R, s must be − R. These values make 1 / s + 1 / R and 1 / s 0 − 1 / R vanish independently so

a( Q ) = 0. Also a( Q ) = 0 when h = 0, for rays incident at the vertex ( s = s 0 = 0) .
1 1 1
( d) For R = 8 cm, = + and s 0 = − ( 1 . 36/ 1 . 70) s = − 0. 8 s
s 0 s 8 cm
Solving these simultaneously gives, s = − 1 8 cm and s 0 = 1 4. 4 cm. Thus aplanatic images occur at
s 0 = 0, 8 cm, and 1 4. 4 cm.
1 06
20-3. The paraxial image point is found from
n1 n2 n2 − n1
+ 0 =
s s R
since s = ∞ ,
1 1 − 1.5
0+ = ⇒ s 0 = 80 cm
s 0 − 40 mm
Then the ray aberration is becomes, ( with s = ∞ , )
( 2 2) 2 2
h 4 n1 1 1 n2 1 1 h 4 n 20 1 1 25 4 1 1 1
a( Q ) = − + + 0 − = − − = − − mm
8 s s R s s0 R 8 s0 s0 R 8 80 80 − 40
a( Q ) = − 0. 858 mm
Proceeding,
2 2
da h 3 n2 1 1 25 3 1 1 1
=− − = − − = − 0. 1 373
dh 2 s0 s0 R 2 80 80 − 40
s 0 da 80 mm
by = = ( − 0. 1 373) = − 1 0. 98 mm
n 2 dh 1
s b y 80
0
bz = = ( − 1 0. 98 mm) = − 35. 1 4 mm
y 25
20-4. Refer to the figure below:
a( Q ) = PQI − POI = ` + ` − ( s + s 0 ) .
In triangle PQC:
Q `2 = R 2 + ( s − R) 2 + 2 R ( s − R) + 2 R ( s − R) cos φ
In triangle CQI:
`
`0 h ` 0 2 = R 2 + ( R − s 0 ) 2 − 2 R ( R − s 0 ) cos φ
R
φ p 1 /2 h h4
O Now cos ϕ = 1 − sin 2 ϕ = 1 − ( h/ R) 2 ≈1− −
P C I F 2R 2 8 R4
s0
R This expression can be used in the expressions for ` and ` 0 and
the resulting expressions expanded. For this task I will ask for
s
help from Maxima:
( %i 1 ) ( F: 1 - h^ 2/ 2/R^ 2 - h^ 4/ 8/R^ 4, di s play( F) )

h2 h4
F=− − +1
2 R2 8 R4
( %i 2) ( L: s qrt( R^ 2+( s - R) ^ 2 +2* R* ( s - R) * F) , di s play( L) )

s
h2 h4
L = R2 + 2 − 2
− 4
+ 1 ( s − R) R + ( s − R) 2
2R 8R
( %i 3) ( Lp: s qrt( R^ 2 +( R- s p) ^ 2- 2* R* ( R- s p) * F) , di s pl ay( Lp) )

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

( R − s) h2 R3 − 2 s R2 + s 3 h4
( %o4) L = s + − + ...
2sR 8 s 3 R3
( %i 5) ( Lpapp: taylor( Lp, h, 0, 4) , ’ Lp Lpapp)

( R − sp) h 2 R 3 − 2 sp R 2 + sp 3 h 4
( %o5) Lp = sp + − + ...
2 sp R 8 sp 3 R 3
The last two expressions are ` and ` 0 to 4 t h order in h

continued . . .
1 07
20-4. continued. . .
Using the expressions from Maxima in the expression for a( Q) gives,

3 2 3
4 R 3
− 2 s 0 2
R + s 03
h4
2 R − 2 s R + s h 0 2
(R − s) h (R − s ) h
a( Q ) = ` + ` 0 − s − s 0 = − + −
2sR 8 s 3 R3 2 s 0R 8 s 03 R3

h2 1 1 2 h4 1 1 1 1 1 1
a( Q ) = + 0− − + − 2 − 2 + +
2 s s R 8 s 3 s 03 s2 R s 02 R R3 R3
Gaussian imaging gives,
1 1 2
+ − =0
s s0 R
So, the term proportional to h 2 vanishes. This relation can be used to eliminate s 0 in the h 4 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. 1 53 cm = 0. 01 53 mm
4 R3 4 ( 400) 3
da h3 25 3
Further, =− 3=− = − 2. 441 × 1 0 − 4
dh s = ∞ R 400 3
Then
s0 da
by = =f = ( 200 cm) − 2. 441 × 1 0 − 4 = − 0. 0488 cm = − 0. 488 mm
n2 dy
s0 f b y 200 ( − 0. 0488)
bz = b y = = cm = − 0. 391 cm = − 3. 91 mm
y h 25

20-6. As in problem 20 5, with s = ∞ , a = − h 4 / 4R 3 . Also, h = f/ ( 2 × 3. 75) = 40 cm
h4 ( 40) 4
( a) a = − = cm = − 0. 00296 cm = − 0. 0296 mm
4R 3 − 4 ( 600) 3
( b) n ∆ t = | a | ⇒ ∆t = | a | / n = ( 0. 0296 mm) / 1 . 4 = 0. 021 mm
20-7. Given: b z = 1 cm, 4 D ( f = 25 cm) , 6 cm diameter lens ( h = 3 cm) . For s = 50 cm, s 0 = s f/ ( s − f) = 50 cm.
da hb (3) ( 1 )
( a) b y = − s 0 = − 0z = − cm = − 0. 6 mm ( b) d b lu r = 2 | b x | = 2 | b y | = 2 ( 0. 6 mm) = 1 . 2 mm
dy s 50
r2 + r 1 − 1 0 + 1 0
20-8. The Coddington shape factor is σ = = = 0 and the focal length is,
r2 − r 1 − 1 0 − 1 0

1 1 1
= (1.5 − 1) − ⇒ f = 1 0 cm
f 1 0 cm − 1 0 cm
For parallel rays, s p0 = f = 1 0 cm. Then, from Eq. ( 20 28)
s 0 − s s/s 0 − 1
p= = = − 1 ( as s → ∞ )
s0 + s s / s 0 + 1
1 1 1 1 1 . 53
0 − = 0 + 0 + [ 3 ( 1 . 5) + 2] [ 1 . 5 − 1 ] ( − 1 ) 2 + ⇒ s h0 = 9. 836 cm
s h 1 0 cm ( 8 ) 1 0 ( 1 . 5) ( 0. 5 ) cm
3 1.5 − 1
b z = s p0 − s h0 = ( 1 0 − 9. 836) cm = 0. 1 64 cm; b y = h/ s 0 ) b z = 0. 01 64 cm
1 08
20-9. The focal length is given by the lensmaker’ s formula

1 1 1
= (1.6 − 1) − ⇒ f = 20 cm = s p0
f 36 cm − 1 8 cm
s0− s r + r 1 − 1 8 + 36
Also, since s → ∞ , p = = − 1 and σ = 2 = = − 1 / 3 . So, using Eq. ( 20 28)
0
s +s r2 − r 1 − 1 8 − 36
( 2 )
1 1 h2 3. 6 1 1 1 . 63
− = − + 4 ( 2. 6) − ( − 1 ) + 6. 8 ( 0. 6) ( − 1 ) +
2
s h0 20 cm 8 ( 20) 3 ( 1 . 6) ( 0. 6) cm 6 3 3 0. 6
1
s h0 =
0. 05/ cm + 0. 00024479 h 2 / cm 3
1
b z = 20 cm − s h0 = 20 cm −
0. 05/ cm + 0. 00024479 h 2 / cm 3
Plotting gives,
3 h ( cm) b z ( mm)
1 0. 974
2.5
2 3. 84
2 3 8. 44
bz (cm)
1.5 4 1 4. 53
1
5 21 . 81
0.5
0
0 1 2 3 4 5 6
h (cm)
r2 + r 1 − 1 0 + 1 0
2 0-1 0. The Coddington shape factor is σ = = = 0 and the focal length is,
r2 − r 1 − 1 0 − 1 0

1 1 1
= (1.5 − 1) − ⇒ f = 1 5 cm
f 1 5 cm − 1 5 cm
s 0 − s 37. 5 − 25
For parallel rays, s p0 = f = 1 0 cm. Then, from Eq. ( 20 28) and with p = 0 = = 0. 2
s + s 37. 5 + 25

1 1 1 1 1 . 53
0 − = 0 + 0 + ( 6. 5) ( 0. 5) ( 0. 2) 2 + ⇒ s h0 = 35. 68 cm
s h 37. 5 cm ( 8 ) 1 5 ( 1 . 5) ( 0. 5 ) cm
3 0. 5
b z = s p0 − s h0 = ( 37. 5 − 35. 68) cm = 1 . 82 cm; b y = h/ s 0 ) b z = ( 2/ 37. 5) ( 1 . 82 cm) = 0. 0970 cm
2 0-1 1 . Since L = ( s p0 − s h0 ) / ( s p0 s h0 ) = b z ( s p0 s h0 ) , a minimum in L gives a minimum in b z . From Eq. ( 20 28) ,

dL h2 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
2 0-1 2 . The equations,

r2 + r1 1 1 1
σ= and = (n − 1) −
r2 − r1 f r1 r2
can be solved simultaneously for r 1 and r2 . The general result is,
2 f (n − 1) 2 f (n − 1)
r1 = and r 2 =
σ+1 σ−1
For σ = 0. 7, n = 1 . 7, and f = 30 cm: r1 = 1 7. 65 cm and r 2 = − 1 00 cm
For σ = 3 , n = 1 . 7, and f = 30 cm: r 1 = 7. 50 cm and r 2 = 1 5. 0 cm
1 09
2 0-1 3. The image position is,
sf ( 30) ( 20
s0= = cm = 60 cm
s − f 30 − 20
The Coddington shape factor is, ( see problem 20 1 1 )
2 ( n2 − 1 ) s 0 − s 2 ( 1 . 6 2 − 1 ) 30 − 60
σ=− =− = 0. 289
n+2 s +s 0 1 . 6 + 2 30 + 60
Then, using the results from problem 20 1 2
2 f ( n − 1 ) 2 ( 20) ( 1 . 6) 2 f ( n − 1 ) 2 ( 20) ( 0. 6)
r1 = = cm = 1 8. 62 cm and r 2 = = = − 33. 75 cm
σ+1 1 . 289 σ−1 − 0. 71 1
2 0-1 4. 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 1 1 )
2 ( n2 − 1 ) s 0 − s 2 ( n2 − 1 ) s 0/ s − 1 2 ( 1 . 62 − 1 )
σ=− = − = − ( − 1 ) = 0. 867
n+ 2 s0+ s n + 2 s 0/ s + 1 1.6 + 2
For light incident on the plane side,
r2 + r 1 r 2 / r 1 + 1
σ= = = − 1 as r1 → ∞
r2 − r 1 r 2 / r 1 − 1
For light incident on the spherical side,
r2 + r 1 1 + r 1 / r2
σ= = = + 1 as r2 → ∞
r2 − r 1 1 − r 1 / r2
Since σ = 1 is closer to 0. 867 than is σ = − 1 , the second orientation is better.
2 0-1 5 . Using the result from problem 20 1 1 ,

2 ( n2 − 1 ) s 0 − s
σ=−
n+ 2 s0+ s
2 ( n2 − 1 ) 2 ( 1 . 52 − 1 )
( a) As s → ∞ , σ → − ( − 1) = ( − 1 ) = 0. 71 4.
n+ 2 1.5 + 2
( b) Using the result found in the solution to problem 20 1 2:
2 f ( n − 1 ) 2 ( 30) ( 1 . 5 − 1 ) 2 f ( n − 1 ) 2 ( 30) ( 1 . 5 − 1 )
r1 = = cm = 1 7. 5 cm and r 2 = = cm = − 1 05 cm
σ+1 0. 71 4 + 1 σ−1 0. 71 4 − 1
2 ( n2 − 1 ) 2 ( 1 . 52 − 1 )
( c) As s 0 → ∞ , σ → − (1) = ( 1 ) = − 0. 71 4. ( Reverse the lens. )
n+2 1.5 + 2

2 n2 − n − 1 s − s0 ( 2 ) 1 . 52 − 1 . 5 − 1
2 0-1 6. ( a) Using Eq. ( 20 31 ) σ = = ( + 1 ) = 0. 8
n+ 1 s + s0 s→∞ 1.5 + 1
( b) Using the result found in the solution to problem 20 1 2:
2 f ( n − 1 ) 2 ( 30) ( 1 . 5 − 1 ) 2 f ( n − 1 ) 2 ( 30) ( 1 . 5 − 1 )
r1 = = cm = 1 6. 7 cm and r 2 = = cm = − 1 50 cm
σ+1 0. 8 + 1 σ−1 0. 8 − 1

2 n2 − n − 1 s − s0 ( 2 ) 1 . 52 − 1 . 5 − 1
( c) σ = = ( − 1 ) = − 0. 8 ( Reverse the lens)
n+1 s + s 0 s 0→∞ 1.5 + 1
2 ( n2 − 1 ) s 0 − s r + r1
2 0-1 7. With s 0 = s , σ = − = 0. So with σ = 2 = 0 ⇒ r 1 = − r 2 . Then,
n+ 2 s0+ s r2 − r1

1 1 1 2
= (n − 1) − = (1.5 − 1) ⇒ r 1 = f = 20 cm, r 2 = − r 1 = − 20 cm
f r1 r2 r1
110
2 0-1 8. In this case,
2 n2 − n − 1 s − s0
σ= = 0
n+ 1 s + s0 s=s0
Thus the answers are the same as in problem 20 1 7.
2 0-1 9. Generally,
X 1 1
=
n i fi R p
For a flat Petzval surface R p → ∞ , so
1 1
+ = 0 ⇒ f2 = − ( n 1 / n 2 ) f1 = − ( 1 . 523/ 1 . 72) ( 20) cm = − 1 7. 71 cm
n 1 f1 n 2 f2
2 0-2 0. ( a) The radius of the Petzval surface is given by Eq. ( 20 32) as,
1 X 1 1 1
= = + ⇒ R p = 1 5. 7 cm
Rp n i 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
2 0-2 1 . 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 = 1 D , V2 = 2 D
fD n1 F − n1 C n2 F − n2 C
r1 1 r2 2
2 − V1 V2 P1 D P2 D
P1 D = PD , P2 D = PD , K1 = , K2 =
1 V2 − V1 V2 − V1 n 1 D − 1 n 2 D − 1
r 1 1 = 2/ K1 , r1 2 = − r 1 1 , r 2 1 = r1 2 , r 2 2 = r 1 2 / ( 1 − K2 r 1 2 )
r1 2 r2 1
1 1 f f
f1 D = , f2 D = , fD = 1 D 2 D
K1 ( n 1 D − 1 ) K2 ( n 2 D − 1 ) f1 D + f2 D
The last line is repeated for fC and fF . A typical Basic program that implements this routine is given
below,
I NPUT 0 0 N1 C, N1 D, N1 F ? 0 0 ; MC, MD, MF
I NPUT 0 0 N2 C, N2D, N2 F ? 0 0 ; NC, ND, NF
I NPUT 0 0 DESI RED F ? 0 0 ; 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)
PRI NT R1 ; R2
PRI NT S1 ; S 2
PRI NT D1 ; D2
PRI NT D
PRI NT C1 ; C2
PRI NT C
PRI NT F1 ; F2
PRI NT F
END
The indices of refraction of the glasses in this problem are given in Table 20 1 as,
nC nD nF
# 51 7/645 1 . 51 461 1 . 51 707 1 . 52262
# 620/380 1 . 61 564 1 . 62045 1 . 631 98
problem 20 21 continued on the next page . . .
111
2 0-2 1 . continued. . .
The step by step calculations proceed as,
n1 D − 1
feq = 20 cm V1 = = 64. 5531
n1 F − n1 C
1 1 n2 D − 1
PD = = D=5D V2 = = 37. 971 2
fin m 0. 2 n2 F − n2 C
− V1 V2
P1 D = PD − 0. 1 21 423 cm − 1 P2 D = PD = − 0. 071 423 cm − 1
V2 − V1 V2 − V1
P1 D P2 D
K1 = = 0. 234830 cm − 1 K2 = = − 0. 1 1 51 5 cm − 1
n1 D − 1 n2 D − 1
2
r1 1 = = 8. 51 68 cm r 1 2 = − r1 1 = − 8. 51 68 cm = r 2 1
K1
r1 2
r2 2 = = − 434. 89 cm
1 − K2 r 1 2
1 1
λ D : f1 D = = 8. 23565 cm f2 D = = − 1 4. 001 0 cm
K1 ( n 1 D − 1 ) K2 ( n 2 D − 1 )
f1 D f2 D
fD = = 20. 000 cm
f1 D + f2 D
1 1
λ C : f1 C = = 8. 2750 cm f2 C = = − 1 4. 1 1 04 cm
K1 ( n 1 C − 1 ) K2 ( n 2 C − 1 )
f1 C f2 C
fC = = 20. 0096 cm
f1 C + f2 C
1 1
λ F : f1 F = = 8. 1 428 cm f2 F = = − 1 3. 7456 cm
K1 ( n 1 F − 1 ) K2 ( n 2 F − 1 )
f1 F f2 F
fF = = 20. 0096 cm
f1 F + f2 F
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.
Prob lems 20 22 and 20 23 are on the next page.
112
2 0-2 2 . 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
# 51 7/645 1 . 63461 1 . 6381 0 1 . 6461 1
# 620/380 1 . 79608 1 . 8051 8 1 . 82771
( a) r1 1 = 3. 4535 cm, r 1 2 = − r 1 1 = − 3. 4535 cm = r2 1 , r2 2 = − 1 2. 6576 cm
f1 ( cm) f2 ( cm) f ( cm)

λ 2. 7061 − 5. 8985 5. 0000
( b) D
λC 2. 721 0 − 5. 9659 5. 0026
λF 2. 6726 − 5. 7380 5. 0026
( c) P1 D = 0. 369534 cm − 1 = 1 / f1 D , P2 D = − 0. 1 69534 cm − 1 = 1 / f2 D .
1 1
Dispersive powers: ∆ 1 = = 0. 01 8022, ∆ 2 = = 0. 039283
V1 V2
( d) Check: V2 P1 D + V1 P2 D = ( 25. 4562 ) ( 0. 369534 cm − 1 ) + ( 55. 48696) ( − 0. 1 69534 cm − 1 ) = 0
2 0-2 3. The calculations proceed as outlined in detail in the solution to problem 20 21 . Data and results are listed
here. feq = − 1 0 cm, PD = − 0. 1 cm − 1 .
nC nD nF
# 573/574 1 . 56956 1 . 57259 1 . 57953
# 689/31 2 1 . 68250 1 . 68893 1 . 70462
( a) r1 1 = − 5. 241 5 cm, r1 2 = − r 1 1 = − 5. 241 5 cm = r2 1 , r 2 2 = 53. 1 840 cm
f1 ( cm) f2 ( cm) f ( cm)

λ D − 4. 5770 8. 4399 − 1 0. 0000
( b)
λ C − 4. 601 3 8. 51 94 − 1 0. 0050
λ F − 4. 5222 8. 2520 − 1 0. 0050
113
C hapter 2 1 Fourier O ptics

mλf 632. 8 × 1 0 − 6 mm × 500 mm
21 -1 . ( a) m λ = d sin θ = Ym d/ f or Ym = =m = 0. 6328 mm
d 0. 5 mm
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. 633 mm
m = 3 , Y3 = 1 . 898 mm
m = 5 , Y5 = 3. 1 64 mm
m 1 d 0. 5 mm
( b) For the fundamental frequency m = 1 so νY = = ⇒λY= = = 0. 5 mm.
d λY m 1

m  m = 1 , KY = 1 2. 57 cycles/ mm
kY 2π 2π
( c) KY = = 2 π νY = 2 π . Thus KY = m= m m = 3 , KY = 37. 70 cycles/ mm
f d d 0. 5 mm  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 = t 1 E1 passes the
first transparency and E2 = t 1 E1 = t 2 t 1 E0 passes the second transparency. Thus the combined transmission
function is t = t 1 t 2 .
( 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 = log 1 0 ( opacity) = log 1 0
T
1
( a) D = log 1 0 = − log 1 0 ( T)
T
( b) As in problem 21 2, T = T1 T2 T3 . Therefore,
D = − log1 0 ( T1 T2 T3 ) = D = − ( log T1 + log T2 + log T3 ) = D 1 + D 2 + D 3
( c) Since the opacity of the i t h transparency is ( opacity) i = 1 . 25, the optical density can be formed as
D i = log 1 0 ( 1 . 25) = 0. 09691
D n et = 5 D i = 0. 48455
Tn et = log − 1 ( − D) = 1 0 − D = 1 0 − . 4 8 4 5 5 = 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) e i k x dx
Substitute for h ( x) the convolution of f and g:
Z Z
F [ h ( x) ] = e i k x 0 0 0
f ( x − x ) g ( x ) d x dx
Changing the order of integration, Z

Z
0
F [ h ( x) ] = g (x ) e ikx
f (x − x ) d x dx0
0
Then let v ≡ x − x 0 so that d x = d v:

Z Z
0
F [ h ( x) ] = g ( x 0 ) e i k ( v + x ) f ( v) d v d x 0
Z Z
0
ikv
F [ h ( x) ] = f( v) e d v g ( x 0 ) e i k x d x 0 or
F [ h ( x) ] = F [ f ( x) ] × F [ g ( x) ]
21 -6. The two functions look like,
u
ρ 1 1 ( u) 6
f( x) f( x − u)
1
−3 3 −6 6
Z
Here, ρ 1 1 ( u) = f( x − u) f( x) d x. As u varies, ρ 1 1 ( u) measures the area of overlap of the square pulses:
For example,
Z3 Z3
2
ρ 1 1 ( 0) = [ f ( x) ] d x = dx = 6
−3 −3
Z3 Z 3
ρ1 1 (3) = f ( x) f ( x − 3 ) d x = 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
Φ 1 1 ( 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
Φ 1 1 ( τ) = A 2 sin β sin ( β + ω τ) d t = A 2 sin β ( sin β cos ω τ + cos β sin ω τ) dτ
− T/ 2 − T/ 2
Z T/ 2 Z T/ 2
Φ 1 1 ( τ) = A 2 cos ω τ sin 2 β d t + A 2 sin ω τ sin β cos β d t
− T/ 2 − T/ 2
even o d d , int = 0
Z T/ 2 Z T/ 2
Φ 1 1 ( τ) = 2 A 2 cos ω τ sin 2 β d t = 2 A 2 cos ω t sin 2 ( ω t + α) dt
0 0
Z T/ 2 T/ 2
2 A 2 cos ω τ 2 A 2 cos ω τ ω t + α sin 2 ( ω t + α
Φ 1 1 ( τ) = sin ( ω t + α) ( ω d t) = −
ω 0 ω 2 4 0

2 A 2 cos ω τ ω T sin 2 ( ω T/ 2 + α) sin 2 α
Φ 1 1 ( τ) = − +
ω 4 4 4
Since ω T = 2 π f T = 2 π,

2 A 2 cos ω τ π sin 2 ( π + α) sin 2 α
Φ 1 1 ( 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 
Φ 1 1 ( τ) = − − +
ω 2 4 4 4
cos ω τ
Φ 1 1 ( τ) = A 2 π
ω
21 -8. ( a) A mirror movement x ( = x 2 − x 1 ) 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 = = = = = 1 8. 3 KHz
∆x λ/ 2 λ 546. 1 × 1 0 − 6 mm
( b) As in measuring a wavelength difference with an interferometer, ( see Eq. ( 8 1 4) ) , the peak to peak dis
tance is
∆x = λ 2ave / 2 ∆ λ
v 2 v ∆λ 2 ( 5 ) ( 5. 97 × 1 0 − 7 )
fb eat = = = Hz = 1 7. 2 Hz
∆x λ 2ave ( 5892. 94 × 1 0 − 7 ) 2
21 -9. If the total mirror movement is 5 cm, then x ω = 2 × 5 = 1 0 cm. According to Eq. ( 21 43) , ∆ λ = λ 2 / x w .
( 6328) 2 ˚ ˚ ( 1 0 4) 2 ˚
( a) ∆λ = A = 0. 040 A ( b) ∆λ = A = 0. 1 00 ˚A
1 0 × 1 08 1 0 × 1 08
116
2 1 -1 0. ( a) During one pass, the total mirror movement is
nm
x = ( 71 . 5 nm/ s ) ( time) = 71 . 5 [ 256 rdg. / ( 1 . 28 rdg. / s) ] = 1 4300 nm
s
x ω = 2 x = 28, 600 nm = 2. 86 × 1 0 3 cm
λ 2 ( 400 × 1 0 − 7 cm) 2
( b) λ m in = = = 5. 59 × 1 0 − 7 cm = 5. 59 nm
xω 2. 86 × 1 0 − 3 cm
2 xω 2 ( 2. 86 × 1 0 − 3
( c) λ m in = = = 2. 24 × 1 0 − 5 cm = 224 nm
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.
λ s h ort est 360 nm
2 1 -1 1 . The given information is,
x w = 2. 78 mm λ: ( 4400 − 400) cm − 1
4400 cm − 1 = 2. 27 µm 400 cm − 1 = 25 µm ( both infrared)
1 1
( a) We want the wave number ν w n = . For resolution we need, ∆ νw n = 2 ∆ λ , where ∆ λ = λ 2 / x ω . Thus,
λ λ

1 λ2 1 1
∆ νw n = = = = 3. 597 cm − 1 ≈ 3. 6 cm − 1
λ2 xω x ω 0. 278 cm

2 x ω 2 ( 2. 78 × 1 0 − 3 m
( b) N − 1 = = = 2449 or N = 2450
λ m in 2. 27 × 1 0 − 6 m
2. 78 mm
( c) V = = 0. 0927 mm/ s
30 s
117
C hapter 2 2 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

F ilm 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 = Bt 1 + B i 1
θt1 B Bb = Bi 2 + Br2 = Bt2
B
E S u b s t rat e
Ea = E0 cos θ 0 − Er1 cosθ 0 = Et1 cos θ t1 − Ei 1 cos θ t1
E
Eb = Ei 2 cos θ t1 − Er2 cos θ t1 = Et2 cos θ t2
B0 Bt1 Bi2 Bt2
n0 n1 ns √
Using B = n ε 0 µ 0 E, the E equations can be expressed in
( a) ( b) terms of B:
B0 cos θ 0 B cos θ 0 Bt1 cos θ t1 B i 1 cos θ t1

√ − r1√
Ea = = √ − √
n 0 ε0 µ 0 n 0 ε 0 µ 0 n1 ε0 µ0 n1 ε0 µ0
B cos θ t1 B r2 cos θ t1 Bt2 cos θ t2
Eb = i 2√ − √ = √
n1 ε0 µ0 n1 ε0 µ0 ns ε0 µ0
√ √ √
n 0 ε0 µ 0 n1 ε0 µ0 n s ε0 µ 0
Define γ0 ≡ , γ1 = , γs =
cos θ 0 cos θ t1 cos θ t2
Then the pair of equations becomes,
B B B B 1 1
Ea = 0 − r1 = t1 − i 1 ⇒ Ea = ( B0 − Br1 ) = ( ( Bt1 − Bi 1 )
γ0 γ0 γ1 γ1 γ0 γ1
B B B 1 1
Eb = i 2 − r2 = t2 ⇒ Eb = ( B i 2 − Br2 ) = Bt 2
γ1 γ1 γs γ1 γs
Now, Bi 2 = Bt 1 e − i δ and Bi 1 = Br2 e − i δ . Using these in the expressions developed for Eb and Bb :
Bt1 e − i δ Bi 1 e i δ Bt 2
B b = Bt1 e − i δ + Bi 1 e i δ = Bt2 Eb = − =
γ1 γ1 γs
B + γ 1 Eb i δ B − γ1 Eb − i δ
Solving these simultaneously for for B i 1 and Bt1 give, Bt1 = b e and Bi 1 = b e
2 2
Substituting these into the expressions for Ea and Ba gives,
B + γ1 Eb i δ Bb − γ1 Eb − i δ
Ba = Bt1 + Bi 1 ⇒ Ba = b e + e
2 2
Bt1 Bi 1 Bb + γ1 Eb i δ B b − γ1 Eb − i δ
Ea = − ⇒ Ea = B t 1 = e − e
γ1 γ1 2 γ1 2 γ1
Simplifying using 2 cos δ = e i δ + e − i δ and 2 i sin δ = e i δ − e − i δ ,
Ba = Bb cos δ + Eb ( i γ1 sin δ)

i sin δ
Ea = B b + 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 = n 1 ε 0 µ 0 / cos θ t 1 while for the case treated in detail in the book leading to Eq. ( 22
√
24) γ1 = n 1 ε 0 µ 0 cos θ t 1 .
118
22 -2 . ( a) The film thickness t should satisfy
λF λ
t= = 0
4 4 n1
√ p
with n 1 = n0 ns = ( 1 ) ( 1 . 5) = 1 . 22. Thus
500 nm
t= = 1 02 nm
4 ( 1 . 22)
( b) The phase difference is

2π 2π
δ = k0 ∆ = ( n 1 t cos θ t1 ) = ( 1 . 22) ( 1 02 nm) ( 1 ) = 1 . 428 rad
λ0 550 nm
So, cos δ = 0. 1 423 and sin δ= 0. 9898. Then,
n 21 ( n 0 − n s ) 2 cos 2 δ + ( n 0 n s − n 21 ) 2 sin 2 δ
R= with n 0 n s − n 21 = 0
n 21 ( n 0 + n s ) 2 cos 2 δ + ( n 0 n s + n 21 ) 2 sin 2 δ
1 . 5 ( 1 − 1 . 5) 2 ( 0. 1 4231 5) 2 + 0
R=
1 . 5 ( 1 + 1 . 5) 2 ( 0. 1 4231 5) 2 + ( 1 . 5 + 1 . 5) 2 ( 0. 989821 ) 2
R = 8. 432 × 1 0 − 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 cos 2 δ = 1
and sin 2 δ = 0,
n2 ( n − ns ) 2 ( 1 ) + 0 ( n0 − ns ) 2
R = 21 0 =
n1 ( n0 + ns ) 2 ( 1 ) + 0 ( n0 + ns ) 2
22 -4. Using Eq. ( 22 4) ,

n 21 ( n 0 − n s ) 2 cos 2 δ + ( n 0 n s − n 21 ) 2 sin 2 δ 1 . 46 2 ( 1 − 1 . 52) 2 cos 2 δ + ( 1 . 52 − 1 . 46 2 ) 2 sin 2 δ
R= =
n 21 ( n 0 + n s ) 2 cos 2 δ + ( n 0 n s + n 21 ) 2 sin 2 δ 1 . 46 2 ( 1 + 1 . 52) 2 cos 2 δ + ( 1 . 52 + 1 . 46 2 ) 2 sin 2 δ
2π 2π 2π
( a) For λ = 800 nm, δ = ∆= ( n 1 t) = ( 1 . 46 ) ( 1 37 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. 031 7 ⇒ R = 3. 1 7%
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/ λ 0 ) ( n 1 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
n 0 n s − n 21 ( 1 ) ( 1 . 52) − ( 2. 35) 2
R= = = 0. 323 ⇒ R = 32. 3%
n 0 n s + n 21 ( 1 ) ( 1 . 52) + ( 2. 35) 2
119
22 -6. The properties of the film for zero reflectance are,
√ p
n 1 = n 0 n s = ( 1 ) ( 4) = 2
λ 2 µm
t= 0 = = 0. 25 µm
4 n1 4 ( 2)
According to Table 22 1 , ZrO 2 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 λ 0 = 550 nm:

n0 = 1 λ0 λ0 550 nm λ0 550 nm
t1 = = = 1 05. 8 nm, t2 = = = 85. 94 nm
n 1 = 1 . 30 4 n1 4 ( 1 . 3) 4 n2 4 ( 1 . 6)
n 2 = 1 . 60
From Eq. ( 22 45) :
n s = 1 . 52
2 2
n 0 n 22 − n s n 21 ( 1 ) ( 1 . 6) 2 − 1 . 52 ( 1 . 3) 2
R= = = 3 × 1 0 − 6 ⇒ R = 0. 0003%
n 0 n 22 + n s n 21 ( 1 ) ( 1 . 6) 2 + 1 . 52 ( 1 . 3) 2
( b) With the layers reversed:
2 2
n 0 n 22 − n s n 21 ( 1 ) ( 1 . 3) 2 − 1 . 52 ( 1 . 6) 2
R= = = 0. 1 555 ⇒ R = 1 5. 55%
n 0 n 22 + n s n 21 ( 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 λ 0 = 550 nm:

n0 = 1 λ 0 = 2 µm
n 1 = 1 . 35λ0 2000
n2 = 2 . 2 t1 =
= nm = 370 nm
4 n 1 4 ( 1 . 35)
n s = 3. 3 λ 2000 nm
t2 = 0 = = 227 nm
4 n2 4 ( 2. 2)
r r
n2 ns 3. 3
( b) The ideal index ratio is = = = 1 . 81 66
n1 n0 1
1 . 81 66 − 1 . 6296
The actual ratio is n 2 / n 1 = 2. 2/ 1 . 35 = 1 . 6296. The % difference is 1 00% = 1 0. 3%
1 . 81 66
( c) The normal reflectance produced is
2 2
n 0 n 22 − n s n 21 ( 1 ) ( 2. 2) 2 − 3. 3 ( 1 . 35) 2
R= = = 0. 01 1 7 ⇒ R = 1 . 1 7%
n 0 n 22 + n s n 21 ( 1 ) ( 2. 2) 2 + 3. 3 ( 1 . 35) 2
22 -9. Refer to Figure 22 5 in the text.
For the λ/ 4 layer: cos δ1 = cos ( π/ 2) = 0, sin δ 1 = 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 λ 0 for which the layers are λ/ 4 and λ/ 2.
1 20
2 2-1 0. 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 ) m1 1 m1 2
M =  ≡ 
i ( γ2 cos δ1 sin δ2 + γ1 sin δ1 cos δ2 ) cos δ1 cos δ2 − ( γ1 / γ2 ) sin δ1 sin δ2 m2 1 m2 2
The general form of the reflection coefficient is, ( since in our case m 1 2 and m 2 1 are pure imaginary)
γ m + γ0 γs m 1 2 − m 2 1 − γs m 2 2
r = 0 11
γ0 m 1 1 + γ0 γs m 1 2 + m 2 1 + γs m 2 2
Now, let us put this result in a form convenient for programming. Since in our case m 1 2 and m 2 1 are pure
imaginary, write m 1 2 = i µ 1 2 , m 2 1 = i µ 2 1 where µ 1 2 and µ 2 1 are real. Then the reflection coefficient can be
recast as,
( γ m − γs m 2 2 ) + i ( γ0 γs µ 1 2 − µ 2 1 )
r = 0 11
( γ0 m 1 1 + γs m 2 2 ) + i ( γ0 γs µ 1 2 + µ 2 1 )
Making the substitutions,
√ √ √ √
γ0 = n 0 ε 0 µ 0 , γ1 = n 1 ε 0 µ 0 , γ2 = n 2 ε 0 µ 0 , and γs = n s ε 0 µ 0 in this expression ( including in the matrix
elements) leads to
( n 0 M − n s P) + i ( n 0 n s N − O ) A + i B
r= ≡
n 0 M + n s P) + i ( n 0 n s N + O ) C+iF
Here
M = cos δ 1 cos δ2 − ( n 2 / n 1 ) sin δ1 sin δ2 P = cos δ1 cos δ2 − ( n 2 / n 1 ) sin δ1 sin δ2
N = cos δ1 sin δ2 / n 2 + sin δ1 cos δ2 / n 1 O = n 2 cos δ1 sin δ2 + n 1 sin δ1 cos δ2
The reflectance is,
2 A2 + B 2
R = | r| =
C2 + F2
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 ( λ 0 / 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 λ 0 / n1 t2 → U = S λ 0 / n2
λ 0 → W ( in loop) δ1 → D δ2 → E
A program to calculate the reflectance for different parameters is given below:
5 PRI NT 0 0 2- LAYER FI LM 0 0
10 I NPUT 0 0 N1 0 0 ; G, 0 0 N2 0 0 ; H, 0 0 NS 0 0 ; I , 0 0 WL0( MI CRON) 00; L
15 I NPUT 0 0 THI CK1 0 0 ; Q, 0 0 THI CK2 0 0 ; 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* SI N( D) * SI N( E) / G
40 N COS ( D) * S I N( E) /H+SI N( D) * CO S( E) / G
45 O G* S I N( D) * COS( E) +H* CO S( D) * SI N( E)
50 P COS ( D) * COS ( E) - G* SI N( D) * SI N( E) / H
55 A M- I * P: B I * N- O : C M+I * P: F I * N+O
60 R 1 00* ( ( A* A+B* B) / ( C* C+F* F) )
65 PRI NT 0 0 WL 0 0 ; W, 0 0 REF 0 0 ; R
70 NEXT W
75 END
Sample output is given on the next page.
1 21
2 2-1 0. Continued. . .
To generate Figure 22 4 in the text, The program output is:
use the following input data:
λ µm R a Rb Rc
Case a Case b Case c 0. 3 7. 84% 4. 91 % 7. 96%
n 1 = 1 . 65 n 1 = 1 . 38 n 1 = 1 . 38 0. 35 1 6. 1 9 4. 48 1 0. 63
n 2 = 2. 1 n 2 = 1 . 6 n 2 = 1 . 85 0. 40 1 4. 02 1 . 80 3. 1 1
Q = 0. 25 Q = 0. 25 Q = 0. 25 0. 45 7. 24 0. 90 0. 01 3
S = 0. 25 S = 0. 25 S = 0. 4 0. 5 1 . 84 1.11 0. 74
In all cases n s = 1 . 52 and λ 0 = 0. 55 µm. 0. 55 0. 1 0 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 1 1 . 6 1 . 22 1.16
2 2-1 1 . 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 ) m1 1 m1 2
M =    =  ≡ 
i γ1 0 i γ2 0 i γ3 0 − iγ1 γ3 / γ2 0 m2 1 m2 2
Using Eq. ( 22 36) ,

γ0 m 1 1 + γ0 γs m 1 2 − m 2 1 − γs m 2 2
r=
γ0 m 1 1 + γ0 γs m 1 2 + m 2 1 + γs m 2 2
γ1 γ3 γ γ γ
γ0 γs [ − i γ2 / ( γ1 γ3 ) ] + i γ1 γ3 / γ2 γ2
− 0γ sγ 2
1 3
r= = γ1 γ3 γ0 γs γ2
γ0 γs [ − i γ2 / ( γ1 γ3 ) ] − i γ1 γ3 / γ2 − γ2
− γ1 γ3
Now r = 0 when
γ1 γ3 γ0 γs γ2
= ⇒ γ0 γs γ22 = γ12 γ32
γ2 γ1 γ3
√
Since for each case γ = m ε 0 µ 0 . Substituting this for each γ and simplifying gives
n 0 n s n 22 = n 21 n 23
n1 n3 √
= n0 ns
n2
2 2-1 2 . 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 MgF 2 for layer 3
so that n 3 = 1 . 35, SiO for layer 2 so that n 2 = 1 . 5, and ZnS for layer 1 so that n 1 = 2. 2. For these choices,
n 1 n 3 ( 2. 2) ( 1 . 35)
= = 1 . 98 ≈ 2
n2 1.5
2 2-1 3. Using Eq. ( 22 53) ,

 2N 2  2N 2
n0 nL 1 1 . 38
 ns nH
−1  1 . 52 2. 6
−1
R = 2N  = 2N 
n0 nL 1 1 . 38
ns nH
+1 1 . 52 2.6
+1
( a) N = 2, R = 81 . 1 % ( b) N = 4, R = 98. 4% ( c) N = 8 , R = 99. 99%
1 22
2 2-1 4. Substituting in Eq. ( 22 53) as in problem 22 1 3, one finds for N = 4, n H = 4, n L = 1 . 35, and n s = 1 . 50.
2 2-1 5. In Eq. ( 22 53) , let x = n 0 / n s and z = n L / n H . Then
√ 1 − x z2 N
± R=
1 + x z2 N
since x < 1 and z < 1 we select the positive square root by writing,
√ 1 − x z2 N
+ R=
1 + x z2 N
Then,
√ √
R + x z2 N R = 1 − x z2N
√ √
x z2 N 1 + R = 1 − R
√
1− R
z2N = 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
2 2-1 6. As in the solution to problem 22 1 5, with x = n 0 / n s and z = n L / n H ,

2
x z2 N − 1
R=
x z2 N + 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
1 23
C hapter 2 3 Fresnel equations
23-1 . From Eq. ( 23 28) , √

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

n1 1 1 1
sin θ c = = ⇒n= = = 1 . 8094
n2 n sin θ c sin ( 33 ◦ 33 0 )
External Reflection:
n2
tan θ p = = n ⇒ θ p = tan − 1 ( 1 . 8094) = 61 ◦ 4 0
n1
Internal Reflection:
n 1 1
tan θ p0 = 1 = ⇒ θ p0 = tan − 1 = 28 ◦ 56 0
n2 n 1 . 8094
23-3. The critical angle is,

θ c = 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 ◦
0
Internal: θ p = tan − 1 ( 1 / n) = tan − 1 ( 1 / 1 . 84) = 28. 5 ◦
23-4. Given,
1
sin θ c ≡ 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 = n 2 / n 1 = n 2
2
1 24
23-5. In each of the 4 equations the desired form results from introducing θ t through the relation n = sin θ/ sin θ t .
For example:
q
√ p
cos θ − n − sin θ cos θ − sin θ 1 / sin θ t − 1
2 2 2 cos θ − sin θ cos 2 θ t ) / sin 2 θ t
rT E = √ = p = q
cos θ + n 2 − sin 2 θ cos θ + sin θ 1 / sin 2 θ t − 1 cos θ + sin θ cos 2 θ t ) / sin 2 θ t
cos θ sin θ t − sin θ cos θ t sin( θ − θ t )
rT E = =−
cos θ sin θ t + sin θ cos θ t sin( θ + θ t )
The other 4 relations are similarly shown to be true. Proceeding by first noting from above that,
√
n 2 − sin 2 θ = sin θ cos θ t / sin θ t
Then,
√ s in 2 θ c os θ sin θ c os θ t
− n 2 cos θ + n 2 − sin 2 θ − sin 2 θ t + sin θ t
rT M = √ = sin 2 θ c os θ sin θ c os θ
n 2 cos θ + n 2 − sin 2 θ + s in θ t
s in 2 θ t t
1 1

− sin θ cos θ + cos θ t sin θ t − sin θ cos θ( sin 2 θ t + cos 2 θ t ) + cos θ t sin θ t sin 2 θ + cos 2 θ
rT M = =
sin θ cos θ + cos θ t sin θ t sin θ cos θ( sin 2 θ t + cos 2 θ t ) + cos θ t sin θ t ( sin 2 θ + cos 2 θ )
− sin θ cos θ sin 2 θ t − sin θ cos θ cos 2 θ t + cos θ t sin θ t sin 2 θ + cos θ t sin θ t cos 2 θ
rT M =
+ sin θ cos θ sin 2 θ t + sin θ cos θ cos 2 θ t + cos θ t sin θ t sin 2 θ + cos θ t sin θ t cos 2 θ
( sin θ cos θ t − cos θ sin θ t ) ( sin θ sin θ t − cos θ cos θ t ) sin( θ − θ t ) cos( θ + θ t )
rT M = =−
( sin θ cos θ t + cos θ sin θ t ) ( sin θ sin θ t + cos θ cos θ t ) sin( θ + θ t ) cos( θ − θ t )
tan( θ − θ t )
rT M = −
tan( θ + θ t )
and,
2 cos θ 2 cos θ 2 cos θ sin θ t 2 cos θ sin θ t
tT E = √ = = =
cos θ + n 2 − sin 2 θ cos θ + sin θ cos θ t / sin θ t cos θ sin θ t + sin θ cos θ t sin ( θ + θ t )
Finally,
2 n cos θ 2 cos θ sin θ/ sin θ t 2 cos θ sin θ t
tT M = √ = =
n 2 cos 2 θ + n 2 − sin 2 θ s in 2 θ c os θ
+
s in θ c os θ t cos θ sinθ + sin θ t cos θ t
sin 2 θ t sin θ t
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 θ t
tT M =
sin( θ + θ t ) cos( θ − θ t )
23-6. Write,
√ √
a 2 + b 2 e − i t an ( b / a )
−1
− n 2 cos θ + i sin 2 θ − n 2 − a + i b a− ib
rT M = √ ≡ =− =− √
a 2 + b 2 e i t an ( b/ a )
−1
n 2 cos θ + i sin 2 θ − n 2 a+ib a+ib
r T M = ( 1 ) e i Φ T M = − 1 e − 2 i t an = e − 2 i t an
−1 −1
( b/ a ) ( b/ a ) + i π
Φ T M = − 2 tan − 1 ( b/ a) + π
π b √ sin 2 θ − n 2
tan − ΦTM = =
2 a n 2 cos θ
√
π sin 2 θ − n 2
tan Φ T M − =−
2 n 2 cos θ
1 25
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.
gnupl ot] n1 2 . 42~ s et ti tl e " External Ref le c ti on" ~ k pi /1 80~ s e t xrange [ 0: 90] ~ s et xzeroaxi s lt 1 ~
s et s i ze 0. 7, 0. 7~ s e t xlabe l " {/ Symbol q}" f ont " Ti mes , 20"
plot ( c os ( k* x) - s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) ) / ( c os ( k* x) +s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) ) ti tle " { /I tal i c
r_{TE} " , ( - n1 * n1 * c os ( k* x) +s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) ) / ( n1 * n1 * c os ( k* x) +s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) )
ti tl e " {/ I tali c r_{ TM}" , 2* c os ( k* x) /( c os ( k* x) +s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) ) ti tl e " {/ I tali c
t_{TE} " , 2 * n1 * c os ( k* x) /( n1 * n1 * c os ( k* x) +s qrt( n1 * * 2 - s i n( k* x) * s i n( k* x) ) ) l w 3 ti tl e " {/ I tali c t_ {TM}"
External Ref lec ti on

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
gnupl ot] n1 1 /2 . 42 ~ k pi /1 80~ s e t xrange [ 0: 30] ~ s et xzeroaxi s lt 1 ~ s e t s ampl es 1 0000
s et s i ze 0. 7, 0. 7~ s e t xlabe l " {/ Symbol q}" f ont " Ti mes , 20" ~ s e t ti tle " I nte rnal Re f le c ti on"
plot ( c os ( k* x) - s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) ) / ( c os ( k* x) +s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) ) ti tle " { /I tal i c
r_{TE} " , ( - n1 * n1 * c os ( k* x) +s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) ) / ( n1 * n1 * c os ( k* x) +s qrt( n1 * * 2- s i n( k* x) * s i n( k* x) ) )
ti tl e " {/ I tali c 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
1 26
23-9. For n = 1 / 1 . 3:
θ c = sin − 1 ( 1 / 1 . 3) = 0. 878 rad = 50. 3 ◦ , θ p0 = tan − 1 ( 1 / 1 . 3) = 0. 6557 = 37. 6 ◦ .
For n = 1 / 2. 42:
θ c = sin − 1 ( 1 / 2. 42) = 0. 426 = 24. 41 ◦ , θ p0 = 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.
GNUpl ot] k pi /1 80~ n1 1 / 2. 42~ s e t xrange [ 0: 90] ~ j p 2 2. 46~ j c 24. 41 ~ s et s ample s 1 00000~
s et s i ze 0. 5, 0. 5~ s e t xlabe l " {/ Symbol q}" f ont " Ti mes , 1 8"
s et ti tle " {/I tali c n} 1 /2 . 42"
s et yti c s ( " - { /S ymbol p} " - pi , " - { /Symbol p} /2" - pi / 2, " 0" 0, " {/S ymbol p/2} " pi / 2, " {/ Symbol p} " pi ) ~ s et
xzeroaxi s l t 3
f 1 ( x) 0< x && x<j p ? 0: 1 / 0
f 2( x) j p< x && x<j c ? pi : 1 / 0
f 3( x) j c < x && x<9 0 ? - 2* atan( s qrt( s i n( k* x) * s i n( k* x) - n1 * * 2) /n1 /n1 /c os ( k* x) ) +pi : 1 /0
g1 ( x) 0< x && x< j c ? 0: 1 /0
g2( x) j c < x && x<90 ? - 2 * atan( s qrt( s i n( k* x) * s i n( k* x) - n1 * * 2 ) / c os ( k* x) ) : 1 /0
plot f 1 ( x) lw 2 noti tle, f 2( x) lt 1 lw 6 noti tle , f 3( x) lt 1 lw 2 ti tl e " {/ Symbol F}_ {TM}" , g1 ( x) lt 2
l w 2 noti tl e, g2 ( x) l t 2 l w 2 ti tle " { /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
2 3-1 0. 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
n 0 n s − n 21 1 . 52 − 1 . 38 2
R= = = 0. 01 26 = 1 . 26%
n 0 n s + n 21 1 . 52 + 1 . 38 2
2 3-1 1 . Using
√ !2 √ !2
cos θ − n 2 − sin 2 θ − n 2 cos θ + n 2 − sin 2 θ
RT E = √ RT M = √
cos θ − n 2 − sin 2 θ n 2 cos θ + n 2 − sin 2 θ
θ RT E RT M
0◦ 2. 01 % 2. 01 %
1 0 ◦ 2. 1 0% 1 . 91 %
45 ◦
5. 23% 0. 274%
90 ◦
1 00% 1 00%
1 27
2 3-1 2 . 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: θ c = sin − 1 ( 1 / 2. 42) = 0. 426 = 24. 41 ◦ , θ p0 = tan − 1 ( 1 / 2. 42) = 0. 392 = 22. 46 ◦ .
TE: θ c = sin − 1 ( 1 / 2. 42) = 0. 426 = 24. 41 ◦ , no Brewster angle.
2 3-1 3. The reflectance and transmission are given by

√ !2 p !2
cos θ − n 2 − sin 2 θ cos ( 50 ◦ ) − 1 . 6 2 − sin 2 ( 50 ◦ )
RT E = √ = p = 0. 1 385 = 1 3. 85% ,
cos θ − n 2 − sin 2 θ cos ( 50 ◦ ) + 1 . 6 2 − sin 2 ( 50 ◦ )
TT E = 1 − R T E = 86. 1 5%
√ !2 p !2
− n 2 cos θ + n 2 − sin 2 θ − 1 . 6 2 cos ( 50 ◦ ) + 1 . 6 2 − sin 2 ( 50 ◦ )
RT M = √ = p = 0. 00623 = 0. 623%
n 2 cos θ + n 2 − sin 2 θ 1 . 6 2 cos ( 50 ◦ ) + 1 . 6 2 − sin 2 ( 50 ◦ )
TT M = 1 − R T M = 99. 38%
2 3-1 4. ( a) For the TE case begin with

E + Er = E t ( 23 20)
n 1 E cos θ − n 1 Er cos θ = n 2 Et cos θ t ( 23 21 )
Solve Eq. ( 23 20) for Er and use the result in Eq. ( 23 21 ) ,
n 1 E cos θ − n 1 ( Et − E) cos θ = n 2 Et cos θ t
Rearrangement gives,
Et 2 n 1 cos θ 2 cos θ 2 cos θ
tT E = = = = √
E n 1 cos θ + n 2 cos θ t cos θ + n cos θ t cos θ + n 2 − sin 2 θ
Similarly for the TM mode

− n 1 E + n 1 E r = − n 2 Et ( 23 22)
E cos θ + Er cos θ = Et cos θ t ( 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 θ t
Et 2 cos θ 2 n cos θ 2 n cos θ
tT M = = = = √
E n cos θ + cos θ t n 2 cos θ + n cos θ t n 2 cos θ + n 2 − sin 2 θ
( b) From Eq. ( 23 20) , after dividing through by E, one finds
1 + rT E = t T E
√ √ √
cos θ − n 2 − sin 2 θ cos θ + n 2 − sin 2 θ + cos θ − n 2 − sin 2 θ 2 cos θ
tT E = 1 + √ = √ = √
cos θ + n − sin θ
2 2 cos θ + n − sin θ
2 2 cos θ + n 2 − sin 2 θ
Similarly after dividing Eq. ( 23 22) by E and by n 1 ,
 − 1 + rT M = − n t T M
√ √ 
1 n cos θ + n − sin 2 θ − − n 2 cos θ + n 2 − sin 2 θ
2 2
1 2 n cos θ
tT M = ( 1 − rT M ) =  √ = √
n n n cos θ + n − sin θ
2 2 2 n cos θ + n 2 − sin 2 θ
2
1 28
2 3-1 5 . ( a) The critical angle is θ c = sin − 1 ( n 2 / n 1 ) = sin − 1 ( 1 / 1 . 458) = 43. 3 ◦
The polarizing angle for external reflection is is θ p = tan − 1 ( n 2 / n 1 ) = tan − 1 ( 1 . 458/ 1 ) = 55. 6 ◦
The polarizing angle for internal reflection is θ p0 = tan − 1 1 / 1 . 458) = 34. 4 ◦

2 2
1−n 1 − 1 . 1 458 2
( b) At normal incidence: R TE = rTE
2
= = = 0. 0347, TTE = 1 − R TE = 0. 9653
1+n 1 + 1 . 1 458 2
√ !2 p !2
cos θ − n 2 − sin 2 θ cos 45 ◦ − ( 1 . 458) 2 − sin 2 45 ◦
At θ = 45 ◦ : R TE = √ = p = 0. 0821
cos θ + n 2 − sin 2 θ cos 45 ◦ + ( 1 . 458) 2 − sin 2 45 ◦
TTE = 1 − R TE = 0. 91 79.
( c) At normal incidence: R TM = R TE = 0. 0347, TTE = 0. 9653
√ !2 √ !2
− n 2 cos θ + n 2 − sin 2 θ − ( 1 . 458) 2 cos 45 ◦ + 1 . 458 2 − sin 2 45 ◦
At θ = 45 ◦ : R TM = √ = √
n 2 cos θ + n 2 − sin 2 θ ( 1 . 458) 2 cos 45 ◦ + 1 . 458 2 − sin 2 45 ◦
R TM = 0. 0067, TTM = 0. 9933
( d) Use Eqs. ( 23 28) and ( 23 29) :
For incident angles less than θ p0 , φ TM = φ TE = 0. So for θ = 0 ◦ and 20 ◦ , φ TM − φ TE = 0.
For θ = 40 ◦ which lies between θ p0 and θ c , φ TM − φ TE = π − 0 = π.
For θ > θ c : with n = 1 / 1 . 458,

√ ! √ !
sin 2 θ − n 2 sin 2 θ − n 2
φ TM − φ TE = − 2 tan − 1 + π + 2 tan − 1
n 2 cos θ cos θ
θ = 50 ◦ : φ TM − φ TE = 2. 43 rad = 1 39 ◦
θ = 70 ◦ φ TM − φ TE = 2. 65 rad = 1 52 ◦
θ = 90 ◦ φ TM − φ TE = π = 1 80 ◦
2 3-1 6. ( a) By trial and error, θ = θ rh om b = 59. 857 ◦ , so that with n = 1 / 1 . 65

√ ! √ !
sin 2 θ − n 2 sin 2 θ − n 2
2 ( φ TE − φ TM ) = − 2 tan −1
+ 2 tan −1
− π = − 1 . 5 π ⇒ π/ 2
cos θ n 2 cos θ
( b) If θ = ( 1 + 0. 05) 59. 857 ◦ = 62. 850 ◦ :

√ ! √ !
sin 2 θ − n 2 sin 2 θ − n 2
2 ( φ TE − φ TM ) = − 2 tan − 1 + 2 tan − 1 − π = − 1 . 543 π ⇒ 0. 457 π
cos θ n 2 cos θ
If θ = ( 1 − 0. 05) 59. 857 ◦ = 56. 864 ◦ :

√ ! √ !
sin 2 θ − n 2 sin 2 θ − n 2
2 ( φ TE − φ TM ) = − 2 tan − 1 + 2 tan − 1 − π = − 1 . 461 π ⇒ 0. 539 π
cos θ n 2 cos θ
1 29
2 3-1 7. Using Eq. ( 23 65) for the TE case:
q
2
cos θ − n 2R − n 2I − sin 2 θ + i ( 2 n R n I

RT E = q
cos θ − n 2R − n 2I − sin 2 θ + i ( 2 n R n I

and for the TM case,

− n 2 − n 2 + i ( 2 n n ) cos θ + n 2 − n 2 − sin 2 θ + i ( 2 n n 2
R I R I
R I

R I

RT M =
n 2R − n 2I + i ( 2 n R n I ) cos θ + n 2R − n 2I − sin 2 θ + i ( 2 n R n I
The given parameters are n R = 2. 485 and n I = 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.
( %i 1 ) ( nR: 2. 484, nI : 1 . 381 , k: %pi /1 80) $
( %i 2) f or j f rom 0 s tep 1 0 thru 9 0 do ( RTE[ j ] : f loat( abs ( ( c os ( k* j ) - s qrt( nR^ 2- nI ^ 2-
s i n( k* j ) * s i n( k* j ) +%i * 2 * nR* nI ) ) / ( c os ( k* j ) +s qrt( nR^ 2- nI ^ 2-
s i n( k* j ) * s i n( k* j ) +%i * 2 * nR* nI ) ) ) ^ 2) , RTM[ j ] : f l oat( abs ( ( ( - nR^ 2 - nI ^ 2+2 * %i * nR* nI ) * c os ( k* j ) +s qrt( nR^ 2 - nI ^ 2 -
s i n( k* j ) * s i n( k* j ) +%i * 2 * nR* nI ) ) / ( ( nR^ 2- nI ^ 2+2 * %i * nR* nI ) * c os ( k* j ) +s qrt( nR^ 2 - nI ^ 2 -
s i n( k* j ) * s i n( k* j ) +%i * 2 * nR* nI ) ) ) ^ 2) )
( %o3) done
( %i 4) f or j f rom 0 s tep 1 0 thru 9 0 do di s play( RTE[ j ] , RTM[ j ] )
θ 0 ◦ 30 ◦ 50 ◦ 70 ◦ 90 ◦
R T E 29. 3 34. 5 45. 4 65. 7 1 00
R T M 29. 3 24. 2 1 4. 9 5. 4 1 00
2 3-1 8. Using the Maxima program from the last problem with the input values n R = 1 . 5 and n I = 5. 3 I find,
θ RT E RT M
0 ◦ 82. 5% 82. 5%
30 ◦ 84. 7% 80. 1 %
60 ◦ 90. 9% 69. 5%
2 3-1 9. Given n I = 5. 3 at λ = 589. 3 nm. ( a) α = 4 π n I / λ = 4 π ( 5. 3) / ( 589. 3 nm) = 0. 1 1 3 nm − 1
( b) I = I0 e − α s . For I = 0. 01 I0 , e − α s = 0. 01 ⇒ s = ( − 1 / α) ln( 0. 01 ) = 40. 75 nm = 0. 069 λ
2 3-2 0. Consider Eq. ( 23 47)

cos θ t
1 = r2 + n t2 n = n2 / n1
cos θ

cos θ t 2
( a) External Reflection: n > 1 ; θ t < θ i ; cos θ t > cos θ. Thus, 1 − = n r2
t > t2 ⇒ t2 < 1 − r2
cos θ
Since r 2 < 1 , t 2 < 1 .
cos θ t 2
Internal Reflection: n < 1 ; θ t > θ i ; cos θ t < cos θ. Thus, 1 − r = n
2
t < t2 ⇒ t2 > 1 − r2
cos θ
No upper limit is imposed.
( c) For the angle of incidence equal to the critical angle, sin θ = sin θ c = n. Then,
2 cos θ 2 cos θ
tT E = √ = =2
cos θ + n 2 − sin 2 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 . . .
1 30
2 3-2 0. 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
2 3-2 1 . ( a) The penetration depth is
1 λ 1 0. 546 µm 1
| z | 1/e = = p = p = 0. 1 64 µm
α 2 π sin θ/ n − 1
2 2 2π sin ( 45 ) / ( 1 / 1 . 6) 2 − 1
2 ◦
1
( b) Since irradiance is proportional to the square of the field amplitude and with α = = 6. 089 µm − 1 ,
| | 1/e
z

I
= e − 2 α | z | = e − 2 ( 6 . 0 8 9 µm
−1 1 µm )
= 5. 1 × 1 0 − 6
I0
1 31
C hapter 2 4 Nonlinear optics and the modulation of light
24-1 . For a single field,

1
E = E0 ( e i ω t + e − i ω t ) = E0 cos ω t
2
The third order polarization is
E03 i 3 ω t ε 0 χ 3 E03
P3 = ε 0 χ 3 E 3 = ε 0 χ 3 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= E0 1 ( e i ω 1 t + e − i ω 1 t ) + E0 2 ( e i ω 2 t + e − i ω 2 t ) ≡ A + B
2 2
P3 = ε 0 χ 3 E 3 = ε 0 χ 3 ( A 3 + B 3 + 3 A 2 B + 3 A B 2 )
Here,
E031 i 3 ω 1 t E3
A3 = e + e − i 3 ω 1 t + 3 e i ω 1 t + 3 e − i ω 1 t = 0 1 ( cos 3 ω1 t + 3 cos ω1 t)
8 4
3
E
B 3 = 0 2 [ cos 3 ω2 t + 3 cos ω2 t]
8
3 E 2
01 E 0 2 i2ω t
3 A2 B = e 1 + e − i 2 ω1 t e i ω2 t + e − i ω2 t
8
3 E021 E0 2 h i ( 2 ω 1 + w 2 ) t i
3 A2 B = e + e − i ( 2 ω1 + ω2 ) t + e i ( 2 ω1 − w2 ) t + e − i ( 2 ω1 − ω2 ) t + 2 ( e i ω2 t + e − i ω2 t)
8
3 E02 1 E0 2
3 A2 B = [ cos 2 ω1 + ω2 ) t + cos ( 2 ω1 − w 2 ) t + 2 cos ω2 t]
4
Similarly, interchanging subscripts 1 and 2,
3 E02 2 E0 1
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 = E0 1 cos ω1 t + E0 2 cosω2 t + E0 3 cos ω3 t
E0 1 i ω 1 t E E
E= e + e − i ω 1 t + 0 2 e i ω 2 t + e − i ω 2 t + 0 3 e i ω 3 t + e − i ω3 t
2 2 2
The second order polarization is P2 = ε 0 χ 2 E 2 . Algebra similar to that of the previous two problems leads
to
ε χ n
P2 = 0 2 ( E021 + E022 + E023 ) + E021 cos 2 ω1 t + E02 2 cos 2 ω2 t + E023 cos 2 ω3 t +
2
+ 2 E0 1 E0 2 [ 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 / n 2 cannot change by reversing E. From Eq. ( 24 7) :
1 1
For + E: = + r ( + E) + R ( + E) 2
n 2 n 20
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.
1 32
λ0 694 nm λ0 1 . 06 µm
24-5. ( a) L C = = = 5893 nm = 8. 62 λ 0 ( b) ∆ n = = = 0. 046
4 ∆ n 4 ( 1 . 534 − 1 . 505) 4 L C 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 × 1 0 − 9 m
VHW = = = 9. 84 kV
2r n 30 2 ( 8. 56 × 1 0 − 1 2 / V) ( 1 . 48) 3
24-7. For this cell,

r 3 r V 30. 9 × 1 0 − 1 2 426
∆n= n 0 E = n 30 = ( 2. 29) 3 = 7. 9 × 1 0 − 6
2 2 L 2 . 01
2π 2π π
Φ= r n 30 V = ( 30. 9 × 1 0 − 1 2 ) ( 2. 29) 3 ( 426) = 1 . 57 or rad
λ0 632. 8 × 1 0 − 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
Combining VHW = and Φ = r n 30 , we have the ratio
2 r n 30 λ0
V Φ λ 0 2 r n 30 Φ
= · =
VH W 2 π r n 30 λ0 π
So,

πΦ Φ
I = Im a x sin 2 = Im ax sin 2
2 π 2
( a) Now I = 0 when the argument of sin 2 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 VH W , 4 VHW , . . . , respectively.
( b) With an initial HWP, the sin 2 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 = VH W , I = Im a x .
24-9. Both effects are present in media lacking inversion symmetry. The ratio of retardations is
Φ p o c k 2 π r n 30 V d2 r n 30 d 2 10 2 3 ( 0. 01 ) 2
= · = = = 73
Φ kerr λ0 2 π K V 2 L K λ 0 VL 1 ( 550 × 1 0 − 9 ) ( 1 × 1 0 4 ) ( 0. 02)
2 4-1 0. The length of the cell is found as,

d d2 ( 0. 01 5) 2
VH W = √ ⇒L= = m = 3. 47 m
2 KL 2 K VHW2 2 ( 0. 036 × 1 0 − 1 2 ) ( 30 × 1 0 3 ) 2
The required cell is too long to be practical.
2 4-1 1 . Starting with the given Doppler effect relation,

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

2ν νS λ λ 1
∆ν= ± = ± ν νS = ± ν νS = ± νS
v 2 v ν
ν 0 − ν = ± νS ⇒ ν 0 = ν ± νS ⇒ ω 0 = ω ± ω S
1 33

W
2 4-1 2 . The distance moved by the sound wave is x = v S t = v S .
v light

nW 1 . 5 ( 0. 01 )
So, x = v S = 3000 = 1 50 nm. The sound wave advances by λ/ 3. 3 for λ = 500 nm.
c 3 × 1 08
2 4-1 3. Adapting the relevant part of Figure 24 1 1 leads to the diagram below,
∆ kS
( a) From the vector diagram, for small ∆ θ , ∆ θ ≈
k
∆k S
2 π 2 π ν S 2 π ∆ν S 2 π
k ( b) k S = = so ∆ k S = . Also, K = .
∆θ λS vS vS λ
k kS ∆ k S 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 = ∆ νS = τ ∆ v S
θ D vS λ vs

D 0. 01
( d) N = ∆ νS = ( 1 20 − 80) 1 0 6 = 67
vS 5. 95 × 1 0 3
2 4-1 4. The required acoustic frequency is found as, λ = λ 0 / n = 2 λ s sin θ = 2 ( v S / νS ) sin θ

2 ( 1 . 6) m
νS = ( 2 n/ λ 0 ) v s sin θ = 2500 sin 1 ◦ = 221 MHz
632. 8 × 1 0 − 9 m s
2 4-1 5 . Using Eq. ( 24 1 8) , λ = ( λ 0 / n) = 2 λ S sin θ and from Snell’ s law sin θ 0 = n sin θ. Therefore,
λ0 λ λ0
λ0
= 2 λ S sin θ 0 / n ⇒ λ 0 = 2 λ S sin θ 0
n
θ θ
kS

λ λ 632. 8 × 1 9 − 9 m
2 4-1 6. Generally, sin θ = = νS = νS = ( 2. 876 × 1 0 − 1 1 s) νS
2 λS 2 vS 2 ( 1 1 × 1 0 3 m/s)
At νS = 50 MHz: sin θ = 0. 001 438; θ = 0. 0824 ◦ = 4. 94 0 At νS = 80 MHz: sin θ = 0. 002301 ; θ = 0. 1 21 8 ◦ = 7. 91 0
So, ∆ θ = 2. 97 0
2 4-1 7. The rotation of the plane of polarization is β = VB d where for a solenoid B = µ 0 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 = = VB d = Vµ 0 n i d ⇒ i d = = A · m = 1 . 59 A · m
4 Vµ 0 n ( 65. 45) ( 4 π × 1 0 − 7 ) ( 6000)
For example for d = 5 cm = 0. 05 m, the necessary current is i = 31 . 8 A.
β 900 min min

2 4-1 8. ( a) At 632. 8 nm, V = = = 0. 0647
B d ( 5098 G) ( 2. 73 cm) G cm
1 330 min min
( b) At 543. 5 nm, V = = 0. 0956
( 5098 G) ( 2. 73 cm) G cm

min
2 4-1 9. β = VB d = 0. 0423 ( 4000 G) ( 5 cm) = 846 0 = 1 4. 1 ◦
G cm
dn V 0. 0423
= = = 0. 071 2 µm − 1
d λ 1 . 0083 λ 1 . 0083 ( 0. 589 µ m)
1 34
C hapter 2 5 O ptical P roperties of Materials
25-1 . ( a) Consider,
K ≡ n 2 = ( n R + i n I ) 2 = n 2R − n 2I + 2 i n I n R = KR + i KI
KR = n 2R − n 2I , KI = 2 n I n R
Solving these two relations for n R and n I proceeds as,

p
KI = 2 n I n R = 2 n I KR2 + n 2I

KI2 = 4 n 2I ( KR2 + n 2I
4 n 4 + 4KR2 n 2I − KI2 = 0
p I p
− 4 KR ± 1 6 KR2 + 1 6 KI2 − KR ± KR2 + KI2
n 2I = =
8 2
To make n 2I > 0, choose the + sign. Thus,
" p # 1 /2
− KR + KR2 + KI2
nI =
2
p p
2 KR − KR + KR2 + KI2 KR + KR2 + KI2
n 2R = KR + n 2I = + =
2 2 2
" p # 1 /2
KR + KR2 + 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 n R , 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
KI n R = =
2 2 ω ε0 2 ε0
1/2
σ µ0 σ
α = 2 KI = = 377
nR ε0 nR
1 35
25-3. Using Eqs. ( 25 37) and ( 25 38) we have,
" # " #
2 2 2
2 2 N e ω 0 − ω Ne2 γω
n R − n I ≡ KR = 1 + 2 n I n R ≡ KI =
m ε 0 ω02 − ω 2 ) 2 + γ 2 ω 2 m ε0 2
ω0 − ω 2 ) 2 + γ 2 ω 2
Solving these simultaneously as in problem 25 1 gives,

" p # 1 /2 " p # 1 /2
KR + KR2 + KI2 − KR + KR2 + KI2
nR = nI =
2 2
To compare to Figure 25 2 use the parameters,
ω0 = 1 0 1 6 s − 1 γ = 1 01 4 s− 1 N = 1 02 8 m− 3
m = 9. 1 09 × 1 0 − 3 1 kg e = 1 . 6 × 1 0− 1 9 C ε 0 = 8. 854 × 1 0 − 1 2 F/ m
Using the free plotting program GNUplot:
Thi s i s a TeXmacs i nterf ac e f or GNUpl ot.

GNUpl ot] w0 1 e1 6~ g 1 e1 4~ N 1 e 28~ eps 0 8. 854e - 1 2 ~ e 1 . 6e - 1 9 ~ m 9. 1 09e- 31
KR( x) 1 +N* e * * 2 /m/eps 0* ( w0* * 2 - x* * 2 ) /( ( w0* * 2- x* * 2) * * 2 +g* * 2 * x* * 2)
KI ( x) N* e * * 2/m/e ps 0* g* x/ ( ( w0* * 2 - x* * 2 ) * * 2+g* * 2* x* * 2)
nR( x) s qrt( ( KR( x) /2 +s qrt( KR( x) * KR( x) +KI ( x) * KI ( x) ) /2 ) ) ~ nI ( x) KI ( x) /2/ nR( x)
s et xrange [ 1 e 1 5: 1 e 1 7] ~ s et s ample s 1 000~ s et yrange [ 0. 001 : 1 0]
s et logs c al e xy~ s et s i ze 0. 6 , 0. 6~ s et xlabel " { /S ymbol w} "
plot nR( x) ti tle " n_R" , nI ( x) ti tl e " 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 × 1 0 2 6 ) ( 2. 70 × 1 0 3 ) − 3
n = 1 electron/ atom N = = m = 6. 027 × 1 0 2 8 m − 3 σ 0 = 3. 54 × 1 0 7 / Ω m
M 26. 982
Ne2 N e 2 ( 6. 027 × 1 0 2 8 ) ( 1 . 602 × 1 0 − 1 9 ) 2 − 1

( a) σ 0 = ⇒ γ= = s = 4. 80 × 1 0 1 3 s − 1
mγ m σ0 ( 9. 1 09 × 1 0 − 3 1 ) ( 3. 54 × 1 0 7 )
N e 2 ( 6. 027 × 1 0 2 8 ) e 2
( b) ω 2p = = = 1 . 91 8 × 1 0 3 2 s − 2 ⇒ ω p = 1 . 38 × 1 0 1 6 s − 1
m ε0 m ε0
( c) At λ = 550 nm, ω = 2 π f = 2 π c/ λ = 3. 425 × 1 0 1 5 Hz. Now,
!
ω2 γ ω 2p
n 2R − n 2I ≡ KR = 1 − 2 p 2 = − 1 5. 347 2 n I n R ≡ KI = = 0. 2291
ω +γ ω ω2 + γ2

" p # 1 /2
KR + KR2 + KI2 KI
nR = = 0. 0292 n I = = 3. 92
2 2 nR
1 36

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 µ 0 = 2 1+ . Use c 2 ≡
c c ω ε0 µ 0

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
ε 0 , 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
δc u = = m = 0. 00856 m = 0. 856 cm
σ µ0 ω ( 5. 76 × 1 0 7 ) ( 4 π × 1 0 − 7 ) ( 2 π) ( 60)
( b) For λ = 3 m
1 /2 1/2
λ 3
δc u = = m = 6. 63 µm
σ µ0 π c ( 5. 76 × 1 0 7 ) ( 4 π × 1 0 − 7 ) π ( 3 × 1 0 8 )
1 /2 1 /2
2 2
25-7. ( a) δ A l = = m = 0. 345 mm
σ µ0 ω 3. 54 × 1 0 7 ( 4 π × 1 0 − 7 ) 2 π × 6 × 1 0 4

3. 54 × 1 0 7
( b) δs . w . = × δA l = 0. 991 m ≈ 1 m
4. 3
1 /2 1 /2
λ 0. 1
25-8. δ A g = = m = 1 . 68 × 1 0 − 6 m = 1 . 68 µm
σ µ0 π c 3 × 1 07( 4 π × 1 0 − 7) π ( 3 × 1 08)
As long as the silver coating is thicker than this the silver plated brass component would work.
25-9. ( a) I = I0 e − α x ⇒ ( I/ I0 ) = ( 1 / 4) = e − α x = e − α ( 3 . 4 2 m ) ⇒ 3. 42 α = ln( 4) ⇒ α = 0. 405 m − 1
( b) ( I/ I0 ) = ( 1 / 1 00) = e − ( 0 . 4 0 5 m
−1
)x
⇒ ( 0. 405 m − 1 ) = ln( 1 00) ⇒ x = 1 1 . 37 m
1 37
2 5-1 0. For a metal we have,
!
ω2 γ ω p2
n 2R − n 2I ≡ KR = 1 − 2 p 2 2 n I n R ≡ KI =
ω +γ ω ω2 + γ2
" p # 1 /2
KR + KR2 + KI2 KI
nR = nI =
2 2 nR
To compare to Figure 25 3 use the parameters,

ω p = 1 . 63 × 1 0 1 6 s − 1 γ = 4. 1 × 0 1 3 s − 1
Using the free plotting program GNUplot:
Thi s i s a TeXmacs i nterf ac e f or GNUpl ot.
GNUpl ot] wp 1 . 6 3e1 6~ g 4. 1 e1 3
KR( x) 1 - wp* wp/ ( x* x+g* g)
KI ( x) g/x* wp* wp/ ( x* x+g* g)
nR( x) s qrt( ( KR( x) /2 +s qrt( KR( x) * KR( x) +KI ( x) * KI ( x) ) /2 ) ) ~ nI ( x) KI ( x) /2/ nR( x)
s et xrange [ 1 e 1 4: 1 e 1 8] ~ s et s ample s 1 000~ s et yrange [ 0. 001 : 1 0]
s et logs c al e xy~ s et s i ze 0. 6 , 0. 6~ s et xlabel " { /S ymbol w} "
plot nR( x) ti tle " n_R" , nI ( x) ti tl e " n_ I "
10
nR
nI
0.1
0.01
0.001
1e+14 1e+15 1e+16 1e+17 1e+18
ω
2 5-1 1 . Referring to page 543,

Ne2 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 ε 0 ω0 ω0 ω0 ω0 ω0 ω0 λ ω0 λ 4
0 0
0 B C 0 0 0
n2 = A + + where A =1+α B = α ( 2 π c) 2 / ω02 C = α ( 2 π c) 4 / ω02
λ2 λ4
" # 1 /2 " ! # 1 /2  !2 
B
0
C
0 0
B /A C /A
0 0 0 √ 0
B /2 A
0
C /2 A 1
0 0 0
8 /A
0
A 1 + 
0 0
n= A + + = A1 / 2 1 + + ≈ + −
λ2 λ4 λ2 λ4 λ2 λ4 8 λ2
( !)
1 B /A
0 0
√ 10 C
0
B2
0
B C
n= A 1 + + 4 0 − 0 = A+ 2 + 4
2 λ2 λ 2A 8A2 λ λ
√ 0
A ≡ A = ( 1 + α) 1 / 2
√ 0 0 2
0 B α ( 2 π c) 2 πc 1
B ≡ B /2 A = = = 2
2 A 2 ω02 ( 1 + α) 1 / 2 ω0 ( 1 + α) 1 / 2
" #
√ 0 C6 0 0
B2 ( 2 π c) 4 α ( 2 π c) 4 α 2 ( 2 π c) 4 α α ( 2 π c) 4 α 4+ 3 α
C≡ A 0 − 0 = 0 − 0 = 0 1− =C=
2A 8A2 2 ω04 A 1 / 2 8 ω04 A 3 / 2 2 ω04 A 1 / 2 4 ( 1 + α) 2 ω04 ( 1 + α) 1 / 2 4 ( 1 + α)
4
πc α ( 4 + 3 α)
C=2
ω0 ( 1 + α) 3 / 2
1 38
2 5-1 2 . 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.
1 39
C hapter 2 6 Laser O peration
26-1 . ( a) According to Eq. ( 26 6) , the spectral energy density has the form
8 π h ν3 1
ρ( ν) =
c3 e h ν / k B T − 1
The energy per unit volume per unit wavelength interval ρ( λ) is related to ρ( ν) by the relation,
Zλ=∞ Zν=∞
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. ( 1 6 6) and changing variables in the frequency integral proceeds as,
Zν=∞ Zν=∞ Zλ=0
8 π h ν3 1 8 πh 1 c
ρ( ν) d ν = d ν = − dλ
c3 e h ν / k B T − 1 λ=∞ λ 3 e h ν/ kBT − 1 λ2
Z νν = 0 ν=0
Z Z
=∞ λ=∞
8 πh 1 c λ=∞
8 πhc 1
ρ( ν) d ν = − 3 h ν/ kBT
− 2
d λ = 5 h ν/ kBT
dλ
ν=0 λ=0 λ e − 1 λ λ=0 λ e −1
So that
8 πh c 1
ρ( λ) =
λ5 e h ν/ kB T − 1
( b) Generally the time averaged energy density h u i of radiation moving in a given direction is related to the
exitance in that direction by the relation M = v h u i . ( 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,
v z = 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
v z , ave = c cosθdω = dϕ cos θ sin θ dθ =
2π 2π 0 0 2
Then the relationship between the energy density and exitance becomes,
1 c
M= h ui
2 2
This relation also holds in each wavelength range so that,
c c 8 πhc 1 2π h c 2 1
Mλ = ρ( λ) = 5
=
4 4 λ e h ν / k B T
−1 λ 5 eh ν/ kBT − 1
1 40
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 h u i = .
∆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 h ui .
∆ A ∆t ∆A ∆z/ c
26-3. Equations ( 26 1 4) ( in steady state) and ( 26 1 6) are,
σI
0 = − A 2 1 N2 − ( N2 − N1 ) (1)
hν 0
NT = N1 + N2 ( 2)
From ( 2) , N1 = NT − N2 . Using this in ( 1 ) and rearranging proceeds as,
σI
0 = − A 2 1 N2 − ( N2 − NT + N2 )
hν 0
N2 A 2 1 + 2 σI/ hν 0 ) = ( σI/ hν 0 ) NT
σI/ ( hν 0 ) NT σI/ ( hν 0 A 2 1 )
N2 = = NT (3)
0
( A 2 1 + 2 σI/ hν ) 1 + 2 σI/ ( hν 0 A 2 1 )
Using ( 3) in ( 2) gives,

1 + 2 σI/ ( hν 0 A 2 1 ) σI/ ( hν 0 A 2 1 ) 1 + σI/ ( hν 0 A 2 1 )
N1 = NT − N2 = − NT =
1 + 2 σI/ ( hν A 2 1 ) 1 + 2 σI/ ( hν 0 A 2 1 )
0 1 + 2 σI/ ( hν 0 A 2 1 )
Then,
σI/ ( hν 0 A 2 1 ) 1 + σI/ ( hν 0 A 2 1 ) NT
Ninv = N2 − N1 = NT − =−
1 + 2 σI/ ( hν 0 A 2 1 ) 1 + 2 σI/ ( hν 0 A 2 1 ) 1 + 2 σI/ ( hν 0 A 2 1 )
26-4. ( a) From Section 26 2,

σNT α0
α = − σ( N2 − N1 ) = ≡
1 + 2 σI/ ( hν 0 A 2 1 ) 1 + 2 σI/ ( hν 0 A 2 1 )
Then α = α 0 / 2 for
hν 0 hν 0
2 σIS , a b s / ( hν 0 A 2 1 ) = 1 ⇒ IS , a b s = A2 1 =
2σ 2στ2
( b) In an ideal four level gain medium the saturation irradiance is given by Eq. ( 26 39) as
hν 0
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 .
1 41
26-5. ( a) The rate equations are given by Eqs ( 26 26) ( 26 29) . Setting the indicated parameters to zero:
dN3 σ I
= − κ 3 2 N3 − p p ( N3 − N0 )
dt hν p
dN2 σI
= κ 3 2 N3 − κ 2 1 N2 − ( N2 − N1 )
dt hν 0
dN1 σI
= κ 2 1 N2 + N − N1 ) − κ 1 0 N1
0 ( 2
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 = − κ 3 2 N3 − ( N3 − N0 ) (1)
hν p
0 = κ 3 2 N3 − κ 2 0 N2 ( 2)
0 = κ 2 1 N2 − κ 1 0 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 × 1 0 3 1 Ip
Ninv = N2 − N1 = ( 5)
1 00000 Ip + 1 . 6565 × 1 0 1 2
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 × 1 0 3 1 Ip
γ0 = σ Ninv = 0. 01 / cm = 1 / m ⇒ Ninv = γ0 / σ = 1 0 2 2 / m 3 =
1 00000 Ip + 1 . 6565 × 1 0 1 2
Numerically I find Ip = 1 1 04 . 4W/ m 2 .
( e) For γ0 = 1 / cm = 1 00/ m. Ninv = 1 0 2 4 / m 3 , Ip = 1 . 1 1 × 1 0 5 W/m 2 .
( f) Solving ( 1 ) ( 4) in steady state for the two pump irradiances, I find,

For Ip = 1 1 04 W/m 2 : For Ip = 1 . 1 × 1 0 5 W/ m 2 :
N0 = 1 . 49990 × 1 0 2 6 / m 3 N0 = 1 . 490 × 1 0 2 6 / m 3
N1 = 1 . 000 × 1 0 1 7 / m 3 N1 = 1 . 000 × 1 0 1 9 / m 3
N2 = 1 . 000 × 1 0 2 2 / m 3 N2 = 1 . 000 × 1 0 2 4 / m 3
N3 = 1 . 000 × 1 0 1 7 / m 3 N3 = 1 . 000 × 1 0 1 9 / m 3
It is reasonable to set N0 ≈ NT in both cases.
( g) Using the suggested relations for an ideal 4 level gain medium

κ 3 2 σ p Ip γ κ h νp 1 1
γ0 = σR p2 τ2 = σ NT τ2 ⇒ Ip = 0 3 = γ0 1 1 04. 3 W/ m
κ 3 h νp σ κ 3 2 σ pτ2 NT 1
For γ0 = 0. 01 / cm = 1 / m; Ip = 1 1 04. 3 W/ m 2 , For γ0 = 1 / cm = 1 / m; Ip = 1 . 1 043 × 1 0 5 W/m 2
1 42
26-6. This is conveniently done with the help of the free computer algebra system Maxima:
( %i 1 ) eq1 : 0 Rp2 - k2 * N2- Q * ( N2- N1 )
( %o1 ) 0 = − ( N 2 − N1 ) Q − k2 N2 + Rp2
( %i 2) eq2: 0 Rp1 +k2 1 * N2- k1 0* N1 +Q* ( N2 - N1 )
( %o2) 0 = ( N2 − N1 ) Q + k2 1 N2 − k1 0 N1 + Rp1
( %i 3) s ols : s olve( [ eq1 , e q2 ] , [ N1 , N2 ] )

( Rp2 + Rp1 ) Q + k2 1 Rp2 + k2 Rp1 ( Rp2 + Rp1 ) Q + k1 0 Rp2
( %o3) N1 = , N2 =
( − k2 1 + k2 + k1 0) Q + k1 0 k2 ( − k2 1 + k2 + k1 0) Q + k1 0 k2
( %i 4) Ni nv: rats i mp( - rhs ( s ols [ 1 ] [ 1 ] ) +rhs ( s ols [ 1 ] [ 2] ) )
( k2 1 − k1 0) Rp2 + k2 Rp1
( %o4) −
( k2 1 − k2 − k1 0) Q − k1 0 k2
Here Q = σI/ hν 0 , so,

( κ 2 1 − κ 1 0 ) R p2 + κ 2 R p1 ( 1 − κ 2 1 / κ 1 0 ) R p2 − ( κ 2 / κ 1 0 ) R p1
N2 − N1 = Ninv = − =
( κ 2 1 − κ 2 − κ 1 0 ) ( σI/ hν 0 ) − κ 1 0 κ 2 κ 2 + ( 1 + κ 2 0 / κ 1 0 ) ( σI/ hν 0 )
26-7.
( a) ( b)
dN3 σ I
3 = − κ 3 2 N3 − p p ( N3 − N1 )
κ3 2 dt hν p
2 dN2 σI
= κ 3 2 N3 − κ 2 1 N2 − ( N2 − N1 )
σIp / hν p κ2 1 σI/ hν 0 dt hν 0

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 = κ 2 3 N3 − p p ( N3 − N1 ) (1)
hν p
0 = κ 3 2 N3 − κ 2 1 N2 ( 2)
NT = N1 + N2 (3)
Solving these, using Maxima, for the given parameters leads to ( in mks units) ,
7. 499925 × 1 0 3 0 Ip − 1 . 242375 × 1 0 3 8
N2 − N1 = Ninv =
50001 Ip + 8. 2825 × 1 0 1 1
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 × 1 0 7 W/ m 2 .
( e) Setting γ0 = 0. 01 / cm = 1 / m = σNinv = ( 1 0 − 2 2 m 2 ) Ninv leads to Ip = 1 . 6567 × 1 0 7 W/ m 2 .

( f) Setting γ0 = 0. 01 / cm = 1 00/ m = σNinv = ( 1 0 − 2 2 m 2 ) Ninv leads to Ip = 1 . 6788 × 1 0 7 W/ m 2 .
( 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.
1 43
26-8. Consider Eq. ( 26 41 ) ,
IL 1
ln + ( IL − I0 ) = γ0 L
I0 IS
If IL IS and I0 IS , this relation becomes,

I
ln L ≈ γ0 L
I0
IL ≈ I0 e γ0 L
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 ZL
dI = γ0 IS dz
I0 0
IL − I0 = γ0 IS L
IL = I0 + γ0 IS L
2 6-1 0. ( a) Using Eq. ( 26 38) and ( 26 39) for ideal 4 level gain media in the expression from problem 26 9 gives,

hν 0
IL − I0 = γ0 IS L = σ R p2 τ2 L = hν 0 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,
Nou t p u t N
= ( IL − I0 ) A/ hν 0 = R p2 A L = R p2 V = p u m 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. )
2 6-1 1 . The relation between the input and output irradiance is,

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

I 1
ln L + ( IL − I0 ) = 2
I0 1 00 W/cm 2
One must solve the relation numerically. I find,
( a) When I0 = 1 W/ cm 2 , IL = 6. 96 W/ cm 2 . ( b) When I0 = 1 0 W/ cm 2 , IL = 49. 7 W/ cm 2
( c) When I0 = 1 00 W/ cm 2 , IL = 221 W/cm 2 ( d) When I0 = 1 000 W/ cm 2 , IL = 1 1 80 W/ cm 2
( e) When I0 = 1 0, 000 W/ cm 2 , IL = 1 0, 200 W/ cm 2
1 44
2 6-1 2 . Using the results from problem 26 1 1 :
( a) IL − I0 = 5. 96 W/ cm 2 , IL / I0 = 6. 96 ( b) IL − I0 = 39. 7 W/ cm 2 , IL / I0 = 4. 97
( c) IL − I0 = 1 21 W/cm 2 , IL / I0 = 2. 21 ( d) IL − I0 = 1 80 W/ cm 2 , IL / I0 = 1 . 1 8
( e) IL − I0 = 200 W/cm 2 , 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.
2 6-1 3. The relation between the input and output irradiance is,

IL 1
ln + ( 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 0 ) and the saturation irradiance is inversely proportional to g( ν 0 ) . So for the case at
hand γ0 ( ν 0 ) = γ0 ( ν0 ) / 2 = 1 / cm and IS ( ν 0 ) = 2IS ( ν0 ) = 200 W/ cm 2 . For these parameters

I 1
ln L + ( IL − I0 ) = 1
I0 200 W/cm 2
One must solve the relation numerically. I find:
( a) When I0 = 1 W/ cm 2 , IL = 2. 70 W/ cm 2 , IL − I0 = 1 . 70 W/ cm 2 , IL / I0 = 2. 70
( b) When I0 = 1 0 W/cm 2 , IL = 25. 2 W/ cm 2 , IL − I0 = 1 5. 2 W/ cm 2 , IL / I0 = 2. 520
( c) When I0 = 1 00 W/ cm 2 , IL = 1 81 W/cm 2 , IL − I0 = 81 W/cm 2 , IL / I0 = 1 . 81
( d) When I0 = 1 000 W/ cm 2 , IL = 1 1 70 W/ cm 2 , IL − I0 = 1 70 W/cm 2 , IL / I0 = 1 . 1 7
( e) When I0 = 1 0, 000 W/ cm 2 , IL = 1 0, 200 W/ cm 2 , IL − I0 = 200 W/ cm 2 , IL / I0 = 1 . 02

2 6-1 4. 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 =
1 0 W/ cm 2 , the middle ( solid) curve is for γ0 = 1 / cm and IS = 20 W/cm 2 . The bottom ( dotted) curve is for
γ0 = 1 / cm and IS = 1 0 W/ cm 2 .
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)
1 45
2 6-1 5. ( a) Use Eq. ( 26 45) with S = R 3 = 1 − T3 . Then

( γ0 − γt h ) L ( γ0 − γt h ) L ( γ0 − γt h ) L
Iou t = T3 IS = T3 IS = T3 IS = IS ( γ0 − γt h ) L
1−S 1 − ( 1 − T3 ) T3
( b) If γ0 γt h ,
Iou t ≈ IS γ0 L
As in problem 26 1 0, using Eq. ( 26 38) and ( 26 39) for ideal 4 level gain media in the expression above

hν 0
Iou t = γ0 IS L = σ R p2 τ2 L = hν 0 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,
Nou t p u t N
= ( IL − I0 ) A/ hν 0 = R p2 A L = R p2 V = p u m 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,
Nou t p u t / t) h ν 0 ν 0
efficiency = = <1
Np u m p / t) hν p νp
1
2 6-1 6. ( a) γt h = ln( 1 / R 3 ) = 0. 0051 3/ cm
L
( γ − γt h ) L ( 2 γt h − γt h ) L 0. 0051 3 × 1 0
( b) Iou t = T3 IS 0 = T3 IS = ( 0. 05) ( 2000) W/ cm 2 = 1 02. 6 W/cm 2
1−S 1 − R3 0. 05
2 6-1 7. The output irradiance has the form,

( γ0 − γt h ) L
Iou t = T3 IS
1−S
It is given that R 3 = 0. 95 = 1 − T3 so A 3 = 0. Further R 1 = R 2 ≡ R. Then
S = R 1 R 2 R 3 = R 2 ( 0. 95) ; γt h L = ln( 1 / S)
So,
( 0. 01 ) ( 1 0) − ln( 1 / ( 0. 95R 2 ) )
2 0. 1 − ln( 1 / ( 0. 95R ) )
2
Iou t = ( 0. 05) ( 2000 W/cm 2 = 1 00 W/ cm
1 − 0. 95 R 2 1 − 0. 95 R 2
Plotting gives,
100
90
80
70
60
Iout (W/cm )
2
50
40
30
20
10
0
0.98 0.985 0.99 0.995 1
R
1 46
2 6-1 8. The output irradiance has the form,

( γ0 − γt h ) L
Iou t = T3 IS
1−S
From the given information, R 3 = 1 − T3 − 0. 01 = 0. 99 − T3 , R 1 = R 2 = 0. 99, γ0 = 0. 01 / cm, IS = 2000 W/ cm 2
and L = 1 0 cm. Then S = 0. 99) 2 ( 0. 99 − T3 )
0. 1 + ln( ( 0. 99) 2 ( 0. 99 − T3 ) )
Iou t = 2000 W/ cm T32
1 − 0. 99) 2 ( 0. 99 − T3 )
The plot looks like,
45 40.8631
40 40.863
35 40.863
30 40.8629
Iout (W/cm2)
Iout (W/cm2)
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.
2 6-1 9. 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 + Iou t as,
1 dI+ γ0 1 dI−
I1 − I2 − = =−
I+ dz 1 + ( I+ + I− ) / IS I− dz
R1 R 2 , 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+ = γ0 dz
I+ dz 1 + ( I+ + C/ I+ ) / IS I+
Z I2 + ZL
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:
√
I1 + I1 − = I2 + I2 − = C ⇒ I1 + ( I1 + / R 1 ) = I2 + ( I2 + R 2 ) ⇒ I1 + = I2 + R 2 R 1 and C = I22 + R 2
Using these in the relations above gives,

1 1 I2 + √ I22 + R 2 1 1
ln + 1 − R1 R2 − − √ = γ0 L
2 R1 R2 IS IS I2 + R 2 R 1 I2 +
1
γ0 L − 2 ln( 1 / ( R 1 R 2 ) TI γ0 ( 2L) − ln( 1 / ( R 1 R 2 )
Iou t = T2 I2 + = T2 IS √ p = 2 S √ p
1 − R1 R2 − R2 + R2 / R1 2 1− R R 1 + R /R 1 2 2 1
1 47
2 6-2 0. When R 1 = 1 , Eq. ( 26 47) becomes,
T2 IS γ0 ( 2 L) − ln( 1 / R 2 ) T I γ ( 2 L) − ln( 1 / R 2 ) T2 IS γ0 ( 2 L) − ln( 1 / S)
Iou t = p √ = 2 S 0 =
2 ( 1 − R2 ) ( 1 + R2 ) 2 1 − R2 2 1−S
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.
2 6-2 1 . a) At line center,

A 2 1 c3 2 h ν0 A 2 1 c 2 1 A 2 1 λ 20 ( 0. 34) 1 0. 6 × 1 0 − 6 ) 2
σ = B 2 1 g( ν0 ) h ν0 / c = = = =
8 π h ν03 π∆νH c 4 π 2 ν02 ∆ νH 4 π 2 ∆ νH 4 π2 ( 1 09)
σ = 9. 69 × 1 0 − 2 2 m 2
( b) Ninv = γ/ σ = ( 3 / m) / ( 9. 69 × 1 0 − 2 2 m 2 ) = 3. 1 × 1 0 2 1 / m 3
h ν0 6. 626 × 1 0 − 3 4 ) ( 3 × 1 0 8 / 1 0. 6 × 1 0 − 6 )
( c) IS ≈ = = 1 . 94 × 1 0 6 W/ m 2
σ τ2 ( 9. 69 × 1 0 − 2 2 ) ( 1 0 × 1 0 − 6 )
2 6-2 2 . The Doppler width is

1 /2 1 /2
8 kB T 8 · 1 . 38 × 1 0 − 2 3 · 400 3 × 1 08
∆ νD = ln( 2) ν0 = ln( 2) = 1 . 5 × 1 0 9 Hz
M c2 ( 20. 2 · 1 . 66 × 1 0 − 2 7 ) 9 × 1 0 1 6 6. 33 × 1 0 − 7
2 6-2 3. 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) e i ω ( 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) e i ω ( t + τ ) = S e i ( ω τ − k P ) E0 ( t) e i ω t
√ − ikP
E0 ( t + τ) = S e E0 ( t)
If the field is resonant with the cavity, e − i k P = 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 − Γ t/ 2
I ∝ E02 ⇒ I = I0 e − 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
1 48
2 6-2 4. Figure 26 1 3 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.
( %i 1 ) ( s c : 3e- 8, d: 1 e- 9, Vr: 0. 3, G: 1 e 8, A: 1 e7, Rth: G* A/s c / Vr, y: 2) $

( %i 2) R: y* Rth$
( %i 3) ( Np[ 1 ] : 0. 0, N2[ 1 ] : 0. 0) $
( %i 4) f or j : 2 thru 1 500 do
( N2[ j ] : N2 [ j - 1 ] +d* ( R- A* N2 [ j - 1 ] - s c * Np[ j - 1 ] * N2[ j - 1 ] ) ,
Np[ j ] : Np[ j - 1 ] +d* ( s c * Vr* ( Np[ j - 1 ] +1 ) * N2[ j - 1 ] - G* Np[ j - 1 ] ) ) $
( %i 5) ( N2l i s t: l i s tarray( N2) , Npli s t: li s tarray( Np) , t: make li s t( d* ( k- 1 ) /1 e- 6 , k, 1 , 1 500) ) $
( %i 6) pl ot2d( [ [ di s c rete , t, N2 li s t] , [ di s c rete, t, 2 0* Npl i s t] ] , [ gnuplot_ c urve _s tyl es , [ " wi th l i ne s 1 " , " wi th li nes
1 " ] ] , [ gnuplot_ pre amble , " s e t nokey; s et s ampl es 2000; s e t xlabe l ’ Ti me ( { /S ymbol m} s ) ’ ; s et yl abel ’ Number
de ns i ty ( c m^ {- 3}) ’ ; s et xrange [ 0: 1 . 5] ; s et labe l 1 ’ {/I tali c N}_ {i nv} ’ at 1 . 2 , 1 . 2e 1 6 ; s e t label 2 ’ { /I tal i c
N_ p} ’ at 1 . 2 , 0. 73e1 6 " ] ) $
( %i 8) pl ot2d( [ [ di s c rete , t, N2 li s t] , [ di s c rete, t, 2 0* Npl i s t] ] , [ gnuplot_ te rm, ps ] , [ gnupl ot_ out_f i l e,
" f i g26_ 1 3. eps " ] , [ gnupl ot_c urve_ s tyle s , [ " wi th li nes 1 " , " wi th l i ne s 2" ] ] , [ gnuplot_ps _term_c ommand, " s et
te rm pos ts c ri pt e ps enhanc e d monoc hrome 1 4" ] , [ gnupl ot_ preambl e, " s et noke y; s et s amples 2 000; s et xl abel
’ Ti me ( {/ Symbol m}s ) ’ ; s e t ylabe l ’ Numbe r dens i ty ( c m^ { - 3}) ’ ; s et xrange [ 0: 1 . 5] ; s e t label 1 ’ { /I tal i c
N} _{ i nv}’ at 0. 73, 1 . 2e 1 6 ; s e t label 2 ’ 2 0{ /I tal i c N_ p}’ at 0. 78, 0. 8e1 6; s et s i ze 0. 7, 0. 7" ] ) ;
2 6-2 5. In figure ( a) , ( b) , and ( d) below the solid curve represents Ninv and the dotted curve is 20 Np. In figure ( c)
the solid curve is Ninv and the dotted curve is 1 00 Np.
7e+16 1.8e+16
1.6e+16
6e+16
1.4e+16
5e+16
1.2e+16
Number dens ty (cm-3)
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
8e+16
Number dens ty (cm )

-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)
2 6-2 6. ( 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.
1 49
2 6-2 7. The pulse width is roughly the inverse of the gain bandwidth ∆ ν = ν ∆ λ/ λ = c ∆λ/ λ 2 .
1 λ2 488 2
Ar + : ∆t p ≈ = = s = 1 . 98 × 1 0 − 1 0 s
∆ν c ∆λ 3 × 1 0 1 7 · 0. 004
1 λ2 633 2
He Ne ∆t p ≈ = = s = 6. 68 × 1 0 − 1 0 s
∆ν c ∆λ 3 × 1 0 1 7 · 0. 002
1 λ2 590 2
Dye ∆t p ≈ = = s = 1 . 45 × 1 0 − 1 4 s
∆ ν c ∆λ 3 × 1 0 1 7 · 80
2 6-2 8. 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 ∼ ( 1 0 + 5 ( 0. 2) − 1 0) Hz = 1 Hz, ∆t p ∼ 1 s, ∆ν ∼ 1 / ∆t p.
( b) ∆ ν ∼ ( 1 0 + 1 0 ( 0. 2) − 1 0) Hz = 2 Hz, ∆ t p ∼ 0. 5, ∆ν ∼ 1 / ∆ t p.
( c) ∆ ν ∼ ( 1 0 + 50 ( 0. 2) − 1 0) Hz = 1 0 Hz, ∆t p ∼ 0. 1 , ∆ ν ∼ 1 / ∆ t p.
2 6-2 9. Let E = Re( Ẽ ) , where,

(N − 1)/2
X
Ẽ = E0 e i ( 2 πν0 t + ϕ 0 ) e i ( 2 πjνfs r t
j= − (N − 1 )/2
For this formula to be sensible N must be odd. Now let x = e i ( 2 π ν fs r t) . Then,

 
(N − 1 )/2
X x( N − 1 ) / 2 + 1 − x− ( N − 1 ) / 2
Ẽ = E0 e i ( 2 πν 0 t + ϕ 0 ) x j  = E0 e i ( 2 πν 0 t + ϕ 0 )
x− 1
j= − (N − 1 )/2
e i ( Nπ ν fs r t) e i πν fs t t − e − i Nπ νf s r t e i πν fs t t i ( 2 πν 0 t + ϕ 0 ) e
i πν fs t t i ( Nπ ν fs r t)
e − e − i Nπ νfs r t
Ẽ = E0 e i ( 2 πν 0 t + ϕ 0 ) = E 0 e
e i 2 πνfs r t − 1 e i πν fs t t e i π νfs t t − e − i π νf s t t
sin( N π νfsr t)
Ẽ = E0 e i ( 2 πν0 t + ϕ 0 )
sin( π νfs r t)
sin( N π νfsr t)
E = Re Ẽ = E0 cos( 2 πν0 t + ϕ 0 )
sin( π νfsr t)
1 50

2 6-30. 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,
sin 2 ( N π νfs r t) sin 2 ( N π νfsr t)
P ∝ E02 ⇒ P = P0
sin ( π νfsr t)
2 sin 2 ( 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,
sin 2 ( N α)
Lim = N2
α → n π sin 2 ( α)
and thus,
Pp = 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 νfs r t 1 = π ⇒ t 1 =
Nνfs r
The width of the pulse from zero to zero on either side of the central maximum is 2 t 1 and so ∆ t p ∼ t 1 is a
reasonable measure of the pulse width. Thus,
1 1
∆tp ∼ =
N νfsr ∆ν
where ∆ ν = N νfsr is the range of frequencies that make up the signal.
2 6-31 . The pulse repetition rate PRR is once per cavity round trip:
c 3 × 1 08
PRR = 1 / ( 2 d/ c) = = Hz = 1 0 8 Hz
2d 2( 1 . 5)
The peak power is
Pave 10 W
Pp = = = 1 430 W
PRR ∆ t p ( 1 0 8 ) ( 70 × 1 0 − 1 2 )
2 6-32 . 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 × 1 0− 7
“vertical spread” ∆θ ≈ = = 1 . 6 rad ≈ 90 ◦
b 1 0− 6
2 λ 2 × 8 × 1 0− 7
“horizontal spread” ∆θ ≈ = = 0. 1 6 rad ≈ 9 ◦
b 1 0 × 1 0− 6
2 6-33. The band gap energy is the energy of the emitted photon,
h c 1 240 eV · nm
E= = = 1 . 55 eV
λ 800
2 6-34. 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 − ( 4 0 ) ( 0 . 0 2 ) = 0. 45
1 51
2 6-35. ( a) In steady state Eqs. ( 26 54) and ( 26 55) become,
0 = R p2 − κ 2 N2 − σc Np N2 (1)
0 = − Γ Np + ( Vg / Vc ) σ c Np N2 ( 2)
From ( 2) ,
Γ Vc
N2 =
σc Vg
Using this in ( 1 ) gives,
2 c κ Γ V
R − κ 2 N2 R p − σc Vg σc ( Vg / Vc ) R p2 − κ 2 Γ
Np = p2 = =
σc N2 Γ Vc / Vg σcΓ
c
( b) In problem 26 23 it is shown that Γ = ( 1 − S) . Also Vg / 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( Np Vc ) c 1 d( Np Vc ) 1
= T3 Vc Np ⇒ Iou t = h ν 0 = h ν 0 T3 ( c/ P) Vc Np = T3 h ν 0 c Np
dt P A dt A
σc ( L/ P) R p2 − κ 2 ( 1 − S) c/ P κ hν 0 ( σR p2 L/ κ 2 ) − ( 1 − S)
Iou t = hν 0 c T3 = T3 2
σc ( 1 − S) c/ P σ 1−S
( c) Now, κ 2 = 1 / τ2 so that
κ 2 hν 0 hν 0 σR p2
= = IS and = σR p2 τ2 = γ0
σ στ2 κ2
Then,
γ0 L − ( 1 − S)
Iou t = 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.
1 52
C hapter 2 7 C haracteristics of Laser Beams
27-1 . ( a) The TEM 0 0 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 TEM 0 0 beam dies off in the radial direction whereas a plane wave represents the ( impossible) situation
for which the wave exists in all space.
( b) TEM 0 0 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 0 0 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,

n 2 ∂ 2 Ẽ
∇ 2 Ẽ − =0
2 c 2 ∂t 2
∂ ∂2 ∂2 k 2 ∂ 2 Ẽ
2
+ 2 + 2 Ẽ − 2 =0
∂x ∂y ∂z ω ∂t 2
Take,
Ẽ = U( x , y , z) e i ( k z − ω t + ϕ )
The necessary x, y and t derivatives are easy to obtain.
∂2 ∂2
2
Ẽ = e i ( k z − ω t + ϕ ) 2 U
∂x ∂x
∂2 ∂2
2
Ẽ = e i ( k z − ω t + ϕ ) 2 U
∂y ∂y
∂ 2 Ẽ
= e i ( k z − ωt + ϕ ) − ω2 U
∂t 2
For z and t a bit more work is required.
∂Ẽ ∂U
= ei( kz − ωt+ ϕ) + i k U ei ( k z − ωt + ϕ)
∂z ∂z

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

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

∂ 2 Ẽ i ( kz − ωt+ ϕ) i ( k z − ω t + ϕ ) ∂U 2 i ( k z − ωt + ϕ) ∂2 U
= e − 2 i ω e − ω U e +
∂t 2 ∂t ∂t 2
Using these all in the wave equation gives
2
∂ U ∂2 U ∂2 U ∂U k2 2
ei( kz − ωt+ ϕ) + + + 2 i k − k 2
− ω U =0
∂x 2 ∂y 2 ∂z 2 ∂z ω2
∂2 U ∂2 U ∂2 U ∂U
+ + +2ik =0
∂x 2 ∂y 2 ∂z 2 ∂z
1 53
27-3. We wish to integrate,
dp i
=
dz z − i z 0
Separation of variables and subsequent integration proceeds as,

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

z − i z0
z
− i ( p( z) − p( 0) ) = ln( z − i z 0 ) | = ln
0
0 = ln( 1 + i z/ z 0 )
− i z0
s
p z 2 + z02 i t an − 1 ( z / z 0 )
1 + z 2 / z 02 e i t an
−1
e − i p( z ) e i p( 0 ) = 1 + i z/ z0 = ( z / z0 )
= e
z 02
s
z 02 − i t an − 1 ( z / z 0 ) i p( 0 )
e i p( z ) = 2 2e e
z + z0
The phase factor e i p( 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 1 4) and ( 27 21 )
π w 02 π ( 5 × 1 0 − 4) 2
q( 0) = − i z 0 = − i =−i m = − i ( 1 . 24 m)
λ 632. 8 × 1 0 − 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 z 0 through the use of Eqs ( 21 29) and ( 27
20) . However it is instructive to note that since z = 50 m z 0 = 1 . 24 m and so R ≈ z. Further the spot size
in the far field is given approximately by,
π
w( z) = z = z θ FF = 50 × 4 × 1 0 − 4 m = 0. 02 m
λ w0
1 1 λ 1 632. 8 × 1 0 − 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 × 1 0 − 4 )
This gives a feel for the nature and appropriateness of the far field approximations.
27-5. ( a) , ( b) Using Eqs. ( 27 1 9) and ( 27 20) together with some of the results from the solution to problem 27 4,
" 2 #
z 02 1 . 24
R( z) = z 1 + 2 = ( 50 m) 1 + = 50. 03 m ≈ z
z 50
r
λz 0
w( z) = 1 + z 2 / z 02 = 0. 0201 6 ≈ z θ FF = 0. 02 m
π
1 54
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 + z 02 / z22 z 2 / z 0 1 + z 02 / z 22 u 2 1 + 1 / u 22
=−1= 2 = ≡ u 1 , 2 ≡ z1 , 2 / z0
RM 1 z1 1 + z 02 / z22 z 1 / z 0 1 + z 02 / z 22 u 1 1 + 1 / u 21
u 1 + u 22
−1= 1 ⇒ u 22 u 1 + u 2 ( 1 + u 21 ) + u 1
u 2 1 + u 21
p
− 1 − u 21 ± ( 1 + u 21 ) 2 − 4 u 21 − 1 − u 21 ± 1 − u 21 )
u2 = = = − u 1 or − 1 / u 1
2 u1 2 u1
For u 2 = − u 1 , z 2 = − z 1 . Then using Eq. ( 27 34) , z2 − z1 = 2 z2 = d ⇒ z 2 = d/ 2, z 1 = − d/ 2. This is the solu

tion of interest.
But what about the other solution? For this case z 1 = − z 02 / z 2 and Eq. ( 27 34) gives

z 2 + z 02 / z 2 = z2 ( 1 + z 02 / z22 = d

But z 2 ( 1 + z 02 / z 22 = R M 2 and since in this case R M 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
RM 1 10
( b) Use, R( z) = R M 2 = z 2 1 + z 02 / z 22 ⇒ z0 = −1 z 12 = − 1 ( 0. 1 7 m) ,
z1 0. 1 7
p p
z 0 = 1 . 293 m = π w 02 / λ ⇒ w 0 = z 0 λ/ π = ( 1 . 293) ( 632. 8 × 1 0 − 9 ) / π = 5. 1 0 × 1 0 − 4 m = 0. 51 0 mm
( c) The spot sizes at the mirrors are equal to each other and each have the value,
p p
w( z 2 ) = w 0 1 + z22 / z 02 = ( 0. 51 0 mm) 1 + 0. 1 7 2 / 1 . 293 2 = 0. 51 4 mm
λ 632. 8 × 1 0 − 9
( d) θ FF = = = 0. 395 mrad
π w 0 π ( 5. 1 0 × 1 0 − 4 )
( e) z FF 50 ( 1 . 293 m) = 64. 6 m
( f) The irradiance at the center of the beam waist is found ( see problem 27 1 3) from,
2 Φ t ot ( 0. 01 W)
I0 = = = 1 . 22 W/cm 2
π w 02 π ( 0. 051 cm) 2
Using Eq. ( 27 26) ,

z 02 1 . 293 2
I( ρ = 0, z = z FF ) = I0 = 1 . 22 W/ cm 2 = 4. 89 × 1 0 − 4 W/cm 2
2 2
( z FF ) + z 0 64. 6 2 + 1 . 293 2
RM 1 RM 2
w( z 2 )
w0 θ FF
z 1 = − d/ 2 z 2 = d/ 2
z=0
1 55
27-7. ( a) The A BCD matrix is,
         
1 0 1 t 1 0 1 0 1 0. 004 1 0 1 . 0007 0. 0027
         
    =     
 n − n0 n   n − n 0 n   0. 5 1 . 5   − 0. 5 1  = 
0 1 0 1 − 0. 531 8 0. 9979
R3 n 0 n0 R2 n 0 n0 − 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. 531 25 1
− 0. 64 1 − 1 . 5( − 2) 1.5
R3
R2
n
n0 n0
27-8. ( a) From the overall matrix in the solution to problem 27 7b,
1
C = − 0. 531 25 = − ⇒ f = 1 . 88 m
f
( b) Using the lensmaker’ s formula for a thin lens,

1 n2 − n1 1 1 1.5 − 1 1 1
= − = − = 1 . 88 m
f n1 R1 R2 1 − 2 m − 0. 64 m
Figure for Problems 27-9 and 2 7-1 0
d `
t
R1 = ∞
R2 R3
q2
q1
1 56
27-9. ( a) The plane mirror has R = ∞ so that.
1 1 λ 1 λ π w 12
= +i = + i ⇒ q 1 = − i = − i z0 1
q1 R 1 π w 12 ∞ π w 12 λ
Note that what is denoted as w 1 here is the spot size at a the beam waist of the beam in the cavity.
( b) Using Eq. ( 27 1 9) , with z = d = 0. 7 m, R = R( z) = 2 m

r r
R 2
R 2 = z 1 + z02 1 / z 2 ⇒ z0 = z − 1 = ( 0. 7 m) − 1 = 0. 954 m
z 0. 7
r r
π w 12 λ z0 1 632. 8 × 1 0 − 9 ( 0. 954)
z0 1 = ⇒ w1 = = = 4. 38 × 1 0 − 4 m = 0. 438 mm
λ π π
( c) Combining the results of parts ( a) and ( b) ,
q 1 = − i z 0 1 = − i ( 0. 954 m)
( d) At the beam waist outside the cavity,
1 C q 1 + D A q 1∗ + B A C q 1 q 1∗ + B C q 1 + A D q 1∗ + B D
= =
q 2 A q 1 + B A q 1∗ + B A 2 q 1 q 1∗ + B 2
Here, in mks units, q 1 = − i ( 0. 954) and,
A = 1 − 0. 53 ` B = 0. 7 + 0. 63 `
C = − 0. 53 D = 0. 63
( e) Using Eq. ( 27 1 7) , R 2 = ∞ and the result from ( d) ,
1 1 λ 1 1 A C q 1 q 1∗ + B C q 1 + A D q 1∗ + B D
= +i = +i =
q2 R2 2
π w2 R 2 z0 2 A 2 q 1 q 1∗ + B 2
The values for the various parameters listed in part ( d) . Equating real parts, ( in mks units)
A C q 1 q 1∗ + B D
0= ⇒ A C q 1 q 1∗ = − B D
A 2 q 1 q 1∗ + B 2
( 1 − 0. 53 `) ( − 0. 53) ( 0. 954) 2 = − ( 0. 7 + 0. 63 ` 0. 63) ⇒ ` = 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 w 22 A 2 q 1 q 1∗ + B 2 λ z0 2
z0 2 = = = 1 . 464 m ⇒ w 2 = = 5. 4 × 1 0 − 4 m = 0. 54 mm
λ ( − B C + A D) | q 1 | π
2 7-1 0. ( 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
1 1 Z1 1 π w0 1 2 1 0. 7 1 π 4. 38 × 1 0 − 4
= 1− + 2 = 1− +
w 02 2 w 02 1 f f λ ( 4. 38 × 1 0 − 4 m) 2 1 . 88 ( 1 . 88 m) 2 632. 8 × 1 0 − 9
w 0 2 = 5. 43 × 1 0 − 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) + z 0 12
( 0. 7 − 1 . 88) 2 + 0. 954 2
These answers correspond reasonably well to the answers found with the more exact and exacting for
malism of the previous solution.
2
( b) These approximate formulas are good only if ( see the discussion following Eq. ( 27 48) ) z 02 1 ( Z1 − f) .
For the case at hand,
z 02 1 = ( 0. 954 m) 2 = 0. 91 0 m 2 , ( Z1 − f) 2 = ( 0. 7 − 1 . 88) 2 = 1 . 3924 m 2
So the validity criterion is not met.
1 57
2 7-1 1 . ( a) The far field distance for the external beam is

zFF = 50 z 0 2 = 50 ( 1 . 464 m = 73. 2 m
Note that z 0 2 was calculated in the solution to problem 27 9e.
λ 632. 8 × 1 0 − 9
( b) The far field divergence angle is, θ FF = = = 3. 73 × 1 0 − 4 rad = 0. 373 mrad
π w 0 2 π ( 5. 4 × 1 0 − 4 )
( c) The spot size at the entrance to the beam expander is,
1 /2 1 /2
w E nt ( z = 30) = w 0 2 1 + ( z/ z 0 2 ) 2 = ( 0. 54 mm) 1 + 30 2 / 1 . 464 2 = 1 1 . 08 mm
Assume that 30 m is in the far field ( it is not according to our adopted criterion but the ratio
z/ z0 2 = ( 30/ 1 . 464) is large enough that the far field formulas are reasonably good. For example using the
far field divergence angle to estimate w E nt gives,

w E nt ≈ z θ FF = ( 30 m) 3. 73 × 1 0 − 4 = 0. 01 1 1 9 m = 1 1 . 1 9 mm
which is a good approximation. The expanded beam has a radius of 1 0 × w E nt ≈ 1 1 1 mm = 1 1 . 1 cm.
( d) The Rayleigh range for the beam exiting the beam expander is
πω02 B E π 0. 1 1 1 ) 2
z0 B E = = m = 6. 1 2 × 1 0 4 m
λ 632. 8 × 1 0 − 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 + z02 B E ( 0. 2 − 0. 1 ) 2 + ( 6. 1 2 × 1 0 4
Using Eq. ( 22 46) ,
2 2 !2
1 1 Z1 1 π w0 1 2 1 0. 2 1 π 1 1 . 1 × 1 0− 2
= 1− + 2 = 2 1 − 0. 1 +
w 02 2 w 02 1 f f λ ( 1 1 . 1 × 1 0 m)
− 2 ( 0. 1 m) 2 632. 8 × 1 0 − 9
!
1 1 π 1 1 . 1 × 1 0− 2
≈ ⇒ w 0 2 = 1 . 81 × 1 0 − 7 m = 0. 1 8 µm
w 02 2 ( 0. 1 m) 2 632. 8 × 1 0 − 9
This spot size is less than λ and so one must question the validity of using the Gaussian beam analysis.
Since z 02 B E Z1 − f) 2 , the approximate formulas would work well.
2 7-1 2 . ( a) , ( b) Equations ( 22 47) and ( 22 46) with Z1 = 0 gives directly,

f2( − f f 2 / z 02 1 f
Z2 = f + 2 2
=f 1− 2 2
= 2/ z2
( − f) + z 0 1 1 + f / z 01 1 + f 01
" 2 #
1 1 1 π w0 1 2 1 1 π w 02 1 1 z 02 1
= + = 1 + = 1 +
w 02 2 w 02 1 f 2 λ w 02 1 f2 λ w 02 1 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 w0 1 / λ
z0 1
f2 λ2
w 041
π2
1 . 01 × 1 0 − 1 5 m 4 w 041
w 0 1 > 1 . 783 × 1 0 − 4 m
So for these parameters as long as w 0 1 is suitable larger than about 0. 2 mm, Z2 ≈ f.
1 58
2 7-1 3. The total power is the irradiance integrated over the transverse plane,
Z 2 Z2π Z∞
w0 2 2
Φ t ot = I dA = I0 dϕ e − 2 ρ / w ( z) ρ d ρ
A w( z) 0 0
Let u = 2 ρ 2 / w 2 ( z) ⇒ d u = 4 ρ dρ/ w 2 ( z) ⇒ ρ dρ = du w 2 ( z) / 4
2 2 Z∞ π
w0 w ( z)
Φ t ot = 2 π I0 e − u du = w 02 I0
w( z) 4 0 2
2 7-1 4. Using the change of variable introduced in the solutions to problem 27 1 3, the fraction of transmitted
power can be formed as
R 2 a2 / w2 − u
Φ ( a) e du 2
/w2
F= = 0R
∞ −u = 1 − e− 2a
Φ t ot 0
e du
2 7-1 5. Use the result of problem 27 1 4 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
/w2
F = 1 − e− 2a
so that,

− 2 a2 / w2 1
e = 1 − F ⇒ − 2 a / w = ln( 1 − F) = − ln
2 2
1−F
√
2a
w= r
1
ln 1 − F
√ √
2 7-1 6. The collimated beam length is taken to be 2 z0 = 2 ( π w 02 / λ) . Also d = 4. 5 ( 2 w 0 ) so w 0 = d/ ( 4. 5 2 ) . Thus,

π d2 π
2 z0 = = d 2 = 1 . 458 × 1 0 − 5 m − 1 d 2
20. 25 λ 20. 25 ( 1 . 064 × 1 0 m
− 6
( 2 z0 ) d = 1 c m = 1 4. 58 m
( 2 z 0 ) d = 2 c m = 58. 3 m
( 2 z 0 ) d = 3 c m = = 1 31 . 2 m
( 2 z0 ) d = 5 c m = 364. 5 m
2 7-1 7. Equation ( 27 58) gives,

2 d
m
Hm ( ξ) = ( − 1 ) m e ξ e − ξ2
dξ m
2 d
0
H0 ( ξ = ( − 1 ) 0 e ξ
2 2 2
e− ξ = ( − 1 ) 0 eξ e− ξ = 1
dξ 0
√
1 ξ2 d − ξ2 ξ2 − ξ2 2 2x
H1 ( ξ) = ( − 1 ) e e =−e − 2 ξe =2 ξ=
dξ w
2
2 ξ2 d −ξ 2
ξ 2 d
−ξ 2
ξ 2
− ξ − 2 ξ − 2 ξ) e − ξ 2
2
H2 ( ξ) = ( − 1 ) e e = e − 2 ξ e = e − 2 e (
dξ 2 dξ !
√
2 x 8 x2
H2 ( ξ) = 4 ξ 2 − 2 = 4 2− 2= 2 − 2
w w
1 59
2 7-1 8. Given,
" 2 #
i λ 2 z 2 + π w 02 λz
p( z) = ln 2 − ( m + n + 1 ) tan −1
2 π w 02 π w 02
Now, z0 = π w 02 / λ, so,
" 2 #
1 z 2 + π w 02 / λ λz
i p( z) = − ln 2 − i( m + n + 1 ) tan −1
2 π w 02 / λ π w 02
2
1 z + z02
i p( z) = − ln − i( m + n + 1 ) tan − 1 ( z/ z 0 )
2 z02
Recall that,

z 2 + z02 z 2 + z02 w 2 ( z)
w 2 ( z) = w 02 1 + z 2 / z02 = w 02 ⇒ =
z 02 z 02 w 02
Then,
2
1 w ( z)
i p( z) = − ln − i( m + n + 1 ) tan − 1 ( z/ z 0 )
2 w 02
" − 1 /2 #
w 2 ( z)
i p( z) + i( m + n + 1 ) tan −1
( z/ z 0 ) = ln
w 02

w0
i p( z) + i( m + n + 1 ) tan −1
( z/ z 0 ) = ln
w( z)
w0
e i p( z ) + i ( m + n + 1 ) t an
−1
( z / z0 )
=
w( z)
w 0 − i ( m + n + 1 ) t a n − 1 ( z / z0 )
e i p( z ) = e
w( z)
2 7-1 9. The row appears as
2
2
Burn Pattern
2
H2 ( x s ) H2 e − x s / 2 H2 e − x s / 2
H20(xs) 2
2 H202 e-xs
H20 e-xs /2
xs
xs
xs
1 60
Φ ( a) 2 2
2 7-2 0. ( a) For the TEM 0 0 mode the fraction transmitted is given in problem 27 1 4 as F0 0 = = 1 − e− 2a /w
Φ t ot
( b) For TEM 0 1 : The irradiance is
2 2 √
I0 1 = I0 w 02 / w 2 H02 ( x s ) H12 ( ys ) e − ρ / w ) =I0 w 02 / w 2 2 2 y/ w e − ρ / w
2 2
Do the integrals in cylindrical coordinates with y = ρ sin ϕ. The fraction transmitted is

R 2π R a
Φ ( a) I0 1 ρ dρ dϕ
F0 1 = = R 02 π R 0∞
Φ t ot I0 1 ρ dρ dϕ 0 0
Further make the change of variables used in problem 27 1 4: u = 2 ρ 2 / w 2 ( z) ⇒ ρ dρ = du w 2 ( z) / 4

After canceling common terms, in the expression for F0 1 ,
R 2π R 2a2/w2
0
sin 2 ϕ dϕ 0 u e − u du
F0 1 = R 2 π R
0
sin 2 ϕ dϕ 0∞ u e − u du
I’ ll use the free computer algebra system,
( %i 1 ) T i ntegrate( ( s i n( j ) ) ^ 2 , j , 0, 2* %pi ) * i ntegrate( u* exp( - u) , u, 0, 2 * a^ 2 /w^ 2) /%pi / i ntegrate( u* e xp( - u) , u, 0, i nf ) ;

Is a zero or nonzero? n on zer o
− 2a2
w2 + 2 a2 e w2
( %o7) F0 1 =1−
w2
2 2
Simplification gives, F0 1 = 1 − 1 + 2 a 2 / w 2 e − 2 a / w
( c) Similarly, for the TEM 1 1 mode with y = ρ sinϕ , x = ρ cosϕ , u = 2 ρ/ w 2

R 2π R ∞ √ 2 √ 2 2 2 R 2π R 2a2 / w2 2 − u
0 0
2 2 y/ w 2 2 x/ w e − 2 ρ / w ρ dρdϕ sin 2 ϕ cos 2 ϕ dϕ 0 u e du
0
F1 1 = R √ 2 √ 2 = R 2π R ∞ 2 −u
2π R ∞ − 2 ρ2 / w 2 sin ϕ cos ϕ dϕ 0 u e du
2 2
0 0
2 2 y/ w 2 2 x/ w e ρ dρ dϕ 0
2!
2 a2 1 2 a2 2 2
Integration under Maxima and subsequent simplification gives, F1 1 = 1 − 1 + 2 + e− 2a /w
w 2 w2
( d) The same procedure gives
R 2π R ∞ √ 2
4 ( 2 y/ w) 2 − 2 e − 2 ρ / w ρ dρ dϕ R 2 π R 2 a / w 1 6u 2 sin 4 ϕ − 8 u sin 2 ϕ + 4 e − u du dϕ
2 2 2 2
0 0 0 0
F0 2 = R √ 2 = R 2π R ∞
2π R ∞ 2− 2 − 2 ρ2 / w 2 1 6u 2 sin 4 ϕ − 8 u sin 2 ϕ + 4) e − u du dϕ
0 0
4 ( 2 y/ w) e ρ dρ dϕ 0 0

1 2 2 2 2
Simplification gives, F0 2 = 1 − 2
2 a /w 2 + 2 a /w + 1 e− 2a /w
2 2
2 3
2 7-2 1 .
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
1 61
2 7-2 2 . From the transmittance curves in problem 27 21 it is evident that the transmittance of the TEM 0 0 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 TEM 0 0 mode has low enough loss to lase.
λ λ 488 × 1 0 − 9
2 7-2 3. ( a) θ FF = ⇒ w0 = = m = 1 . 55 × 1 0 − 4 m = 0. 1 55 mm
π w0 π θ FF π ( 0. 001 )
2 Φ t ot 10
( b) I0 = = W/m 2 = 1 . 32 × 1 0 8 W/ m 2 = 1 . 32 × 1 0 4 W/ cm 2
π w 02 π ( 1 . 55 × 1 0 − 4 ) 2
2
π w 02 π 1 . 55 × 1 0 − 4
( c) The Rayleigh range is z 0 = = m = 0. 1 55 m. The irradiance is then
λ 488 × 1 0 − 9
z 02 0. 1 55 2
I( ρ = 0, z = 1 0 m) = I0 = 2 1 . 32 × 1 0 4 W/ cm 2 = 3. 1 7 W/ cm 2
z2 + z0 2 1 0 + 0. 1 55 2
2 7-2 4. The Rayleigh range is needed:

The beam waist is, by symmetry, in the center of the cavity and so the z 1 = − 0. 1 m and z 2 = 0. 1 m. Also
R 2 = 1 m. Then,
r
z 02 R( z 2 )
R( z 2 ) = z 2 1 + 2 ⇒ z0 = z 2 − 1 = 0. 3 m
z2 z2
Equation ( 27 67) gives the cavity mode frequencies as

m+ n+ 1
νm n q = q + tan − 1 ( z 2 / z0 ) − tan − 1 ( z 1 / z 0 )
π
Using this relation,
3 × 1 08
( a) ν0 , 0 , q + 1 − ν0 , 0 , q = c/ 2d = Hz = 750 MHz
0. 4
3 × 1 08
( 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 = 1 54 MHz
2d π z z
0 0
c 1 z z
( d) ν 1 , 0 , q − ν0 , 0 , q = tan − 1 2 − tan − 1 1 = 1 54 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
1 62
C hapter 2 8 S elected 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 . 06 µm lens
Pulse width: 1 0 ns
Energy per pulse: 4 mJ w0 1 w0 2
Spot size of beam
incident on focusing 3. 4 mm focal sp ot
lens
Power of focusing lens 20 D iris
f
Assume that the beam is incident on the lens at a waist of radius w 0 1 = 3. 4 mm. The Eq. ( 27 46) ( with
Z1 =0) give the focal spot radius as,
!
1 1 1 π w0 1 1 1 π ( 3. 4 × 1 0 − 3 m
= + = + ⇒ w 0 2 = 5 µm
w 02 2 w 02 1 f 2 λ ( 3. 4 × 1 0 − 3 ) 2 ( 0. 05 m) 2 1 . 06 × 1 0 − 6 m
The confocal parameter for the focused beam is,
π ( 5 × 1 0 − 6) 2
z 0 2 = π w 02 2 / λ = m = 7. 40 × 1 0 − 5 m = 0. 074 mm
1 . 06 × 1 0 − 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 × 1 0 5 W
1 0− 8 s
For a TEM 0 0 beam, the irradiance at the center of the spot is given by problem 27 1 3 as,
2Φ 8 × 1 05
I= 2 = W/ m 2 = 1 . 0 × 1 0 1 6 W/m 2 = 1 0 1 2 W/cm 2
π w0 2 π ( 5 × 1 0 − 6 ) 2
This value is in rough agreement with the estimate given in Example 28 1 .
r r
ε0 c 2 2I 2 · 5. 7 × 1 0 1 5
28-5. I = E ⇒ E0 = = = 2. 1 × 1 0 9 V/ m
2 0 ε0 c 8. 85 × 1 0 − 1 2 · 3 × 1 0 8
28-6. Using Eq. ( 27 48) one can estimate the beam diameter on the earth as,

D = f ( 2 θ FF ) = ( 3. 6205 × 1 0 7 m 5 × 1 0 − 6 = 1 80 m

The average irradiance over the beam is found roughly as, I = Φ / ( π D 2 / 4 = 7700 W/ m 2 . Using the result
from problem 28 5,
r r
ε c 2I 2 · 7700
E0 = I = 0 E02 ⇒ E0 = = V/m = 2400 V/ m
2 ε0 c 8. 85 × 1 0 − 1 2 · 3 × 1 0 8
1 63
28-7. ( a) , ( b)
" 1 /2 #
1 4P 1 4 × 50
R NH Z = −d = − 0. 3 cm
d ϕ π MPE 0. 003 π 5. 1 × 1 0 − 4
Laser ϕ
R NH Z = 37, 1 00 cm = 371 m
R NH Z
28-8. ( a) , ( b)
" 1 /2 #
f 4P 7. 5 4 × 50
R NHZ = = cm
d b π MPE 0. 5 π 1 . 6 × 1 0− 3
b ϕ ≈ b/ f
Laser
R NHZ = 2990 cm = 29. 9 m
f R NHZ
1 1 1 1 1
2 8-1 0. ( a) ν p = νs + νi ⇒ νi = ν p − νs ⇒ = − = − ⇒ λ i = 361 0 nm
λ i λ p λ s 1 060 nm 1 500 nm
( b) The following two relations must both be satisfied,
d = ms λ s / 2 = mi λ i / 2
d≈ 1 m
with d being the cavity length. These conditions are met for, for example, by

d = 1 . 001 775 m = 370 ( 361 0) ( 1 500 × 1 0 − 9 m / 2 = 370 ( 1 500) ( 361 0 nm) / 2
In practice one would need active control of the cavity length to maintain the dual resonance.
( c) If both beams are TEM 0 0 modes the resonant frequencies are given by Eq. ( 27 67) as,

1 c
ν0 0 q = q + tan − 1 ( z2 / z 0 ) − tan − 1 ( z 1 / z 0 )
π 2d

c 1 λ 00 q tan − 1 ( z 2 / z 0 )
d= q+ tan ( z2 / z 0 ) − tan ( z1 / z 0 ) =
−1 −1
q+
2 ν0 0 q π 2 π
Here I have noted that the beam waist is at the flat mirror at z 1 = 0. The confocal parameter can be found
from the relation,
R 2 = R( z2 = d) = d ( 1 + z 02 / d 2 )
− 1 /2
R2
−1 = d/ z0 = z 2 / z 0
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 ( z 2 / z 0 ) 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. 9991 366875 m = 369 [ ( 361 0) + 1 / 4] ( 1 500 × 1 0 − 9 m / 2 = 369 [ ( 1 500) + 1 / 4] ( 361 0 × 1 0 − 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.
1 64
2 8-1 1 . 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
Pt an sm it t ed Pin c 0. 1 W
Pc oprop = = = = 3. 33 W
T ( 1 − R) 1 − 0. 97
The counter propagating intracavity field will have a power of,

Pc ou nt erp rop = R Pc op rop = ( 0. 97 ( 3. 33 W) = 3. 23 W
r
3 kB T
2 8-1 6. From thermodynamics: v rm s = .
m r
3 · 1 . 38 × 1 0 − 2 3 · 293
( a) For the oxygen molecule at T = 68 F = 20 C = 293 K, v rm s ≈
◦ ◦
m/s = 480 m/ s
32 · 1 . 67 × 1 0 − 2 7
r r
3 kB T 3 · 1 . 38 × 1 0 − 2 3 · 0. 002
( b) For rubidium at T = 0. 002 K, v rm s = ≈ m/ s = 0. 8 m/s
m 85. 5 · 1 . 67 × 1 0 − 2 7
2 8-1 7. As in the solution to problem 28 1 6b but with T = 0. 003 K,

r r
3 kB T 3 · 1 . 38 × 1 0 − 2 3 · 0. 003
vr m s = ≈ m/ s = 0. 98 m/ s
m 85. 5 · 1 . 67 × 1 0 − 2 7
The detuning should match the Doppler shift. So,

∆ν ≈ ν v rm s / c = v rm s / λ = ( 0. 98 m/s) / 800 × 1 0 − 9 m = 1 . 225 × 1 0 6 Hz
1 65

Manual Introduction To Optics Pedrotti PDF

Uploaded by

Copyright:

Available Formats

Manual Introduction To Optics Pedrotti PDF

Uploaded by

Document Information

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

Manual Introduction To Optics Pedrotti PDF

Uploaded by

Copyright:

Available Formats

C hapter 1 Nature of Light

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

1 -1 7. a) The half angle divergence θ 1 / 2 can be found from the relation

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

h1 The person must be able to see the top of

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,

One finds s 1 = 1 1 2. 5 cm, s 2 = 1 00 cm, s 10 = 225 cm, s 20 = 300 cm, f = 75 cm

For the negative meniscus lens

For the back focal length: for lens 1 :

2-2 3. The arrangement of the object and lenses is shown below.

2-2 7. 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,

For the situation in which the center of curvature in medium with n = 4/ 3:

The image distance is then found as,

The line image is real, 1 8. 75 cm past the lens and 1 5. 75 cm long.

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.

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

( d) Exit Pupil E x 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

( a) The Cauchy relations for λ F and λ C are: n F = A + B/ λ 2F , n C = A + B/ λ 2C . Solving these simultaneously,

4240 1 . 5286 − 1 . 5205

Then for a third wavelength λ D = 589. 2 nm,

( c) The resolving powers are,

The minimum resolvable wavelengths are,

crown: ( ∆λ) m in = λ/ R C = 5892 ˚A / 31 1 0 = 1 . 9A

3-1 0. The indices of refraction of the lines are,

δF 2 = ( 1 . 536 − 1 ) ( 5. 99) = 3. 21 1 ◦ δF 1 = ( 1 . 648 − 1 ) ( 5 ) = 3. 240 ◦ Net: δF 2 − δF 1 = 0. 029 ◦

Dispersion = δF 1 2 − δC 1 2 = 0. 029 − ( − 0. 007) = 0. 036 ◦ = 2. 1 6 0

The relative dispersion is δF 1 − λ C 1 = 0. 1 21 5 ◦ . This must be reversed by the flint prism:

The deviation of line D is then, δD 1 − δ D 2 = ( n D 1 − 1 ) A 2 = ( . 523) ( 1 5) − ( 0. 7205) ( 4. 821 ) = 4. 37 ◦

3-1 4. Plots as a function of the different parameters are shown below.

3-1 5. From Eqs. ( 3 30) and ( 3 31 ) ,

3-1 9. ( a) Using f = 1 5 cm, s = 1 00 ft = 3048 cm, and h = 6 ft, one finds,

The ratio of the magnifications is

3-2 3. ( a) The system is sketched below.

( b) When viewed at ∞ , M = fo / fe = 30/ 4 = 7. 5 × . When viewed at 25 cm, s 00 = − 25 cm and so,

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

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

The overall magnification is then, Moverall = m M = ( 5) ( − 36) = − 1 80 ×

4-1 . The waveform is sketched below. As a function of time y = A e − b ( x + 1 0 t)

Wave form 3 is not a traveling wave since it is not a function of simply u = x − v t.

4-4. The waveform can be rewritten as,

4-5. From the given information, k = 2 π/ λ = ( 2 π/ 5) m 

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

4-1 2 . Let z̃ = a + i b where a and b are real.

4-1 3. ( a) Note that e i π/ 2 = cos( π/ 2) + i sin( π/ 2) = i so that

4-1 4. One can write ψ = Im( ψ̃ ) with

ψ = Im( ψ̃ ) = 2 A cos( k y) sin( ω t)

4-1 7. ( a) From the solution to problem 4 1 5,

4-2 0. Eq. ( 4 44) gives,

4-5. From the given information, k = 2 π/ λ = ( 2 π/ 5) m

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

7-2 4. Using Eq. ( 7 39) with t m rm ,

7-2 5. Using Eq. ( 7 39) with t m rm ,

1 0-1 8. The group delay is