0% found this document useful (0 votes)
21 views37 pages

Chapter 1

This document outlines the quantization of radiation fields by: 1) Deriving Maxwell's equations in terms of vector and scalar potentials in the Coulomb gauge. 2) Expressing the vector potential as a sum of positive and negative frequency parts, each satisfying the wave equation. 3) Showing the positive frequency part satisfies the wave equation and Coulomb gauge condition, resulting in a second-order differential equation for its dynamical variable.

Uploaded by

teshomedebela
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
21 views37 pages

Chapter 1

This document outlines the quantization of radiation fields by: 1) Deriving Maxwell's equations in terms of vector and scalar potentials in the Coulomb gauge. 2) Expressing the vector potential as a sum of positive and negative frequency parts, each satisfying the wave equation. 3) Showing the positive frequency part satisfies the wave equation and Coulomb gauge condition, resulting in a second-order differential equation for its dynamical variable.

Uploaded by

teshomedebela
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 37

Home Page

The Quantum Description of


Title Page Radiation
Contents

JJ II

J I
Lecture Notes
Page 1 of 37

Go Back

Full Screen Interaction of Light with Condensed Matter (Phys 843)


Close

Quit Department of Physics, AAU


Addis Ababa, Ethiopia

October 24, 2023


OUTLINE

1. Quantization of the Radiation Fields

Home Page 2. The Number State


Title Page
3. The Coherent State
Contents
4. The Chaotic State
JJ II

J I

Page 2 of 37

Go Back

Full Screen

Close

Quit
1. Quantization of the Radiation Fields

• The Maxwell’s equations that describes the electromagnetic fields in


free space (SI units) have the form
Home Page

Title Page
~ = 0,
∇·E (1)

Contents

~
JJ II ~ = − ∂B ,
∇×E (2)
∂t
J I

Page 3 of 37
~ = 0,
∇·B (3)

Go Back
~
Full Screen ∇×B ~ = 1 ∂E . (4)
c2 ∂t
Close ~ is the electric field, B
where E ~ is the magnetic field, and c is the

Quit
speed of light in vacuum.

• From vector calculus, we know that the divergence of curl of an ar-


bitrary vector F~ , vanishes. That is,

∇ · (∇ × F~ ) = 0. (5)
• In terms of the vector potential, the magnetic field can be expressed
as
~ = ∇ × A,
B ~ (6)

Home Page
~ r, t) is the vector potential. Using (5) in (2), we obtain
where A(~

Title Page ~
∂(∇ × A)

∂A~
∇×E =− ~ , ⇒ ∇× E+ ~ = 0. (7)
Contents
∂t ∂t

JJ II • Using the identity ∇ × ∇φ = 0 the electric field is written as


J I ~
~ = − ∂ A − ∇φ,
E (8)
Page 4 of 37 ∂t

Go Back
where φ = φ(~r, t) is the scalar potential.

Full Screen • Substituting (8) into (1), we get


 ~ 
Close ∂A ∂ ~
∇· + ∇φ = 0, ⇒ ∇2 φ + (∇ · A). (9)
Quit
∂t ∂t

• The quantum theory of radiation is formulated using the Coulomb


gauge, which is defined by

~ = 0.
∇·A (10)
• Using the Coulomb gauge in (9), we obtain the Laplace equation to
be
∇2 φ = 0. (11)

Home Page • Suppose the scalar potential is a function of position, r, only; i.e.,
Title Page φ(r, θ, ϕ) = φ(r) in spherical coordinates. Then, the general solution
of the Laplace equation takes the following form:
Contents
 
JJ II 2 1 d 1 dφ c1
∇ φ(r) = 2 2
= 0, ⇒ φ(r) = − + c2 (12)
r dr r dr r
J I
where c1 and c2 are constants to be determined from the boundary
Page 5 of 37
(physical) conditions. The boundary conditions must be finite (zero,
Go Back for convenience) for (i) r → ∞ and (ii) r → 0. Employing these
Full Screen conditions, we find that

Close 1. r → ∞ ⇒ c1 = 0.
Quit 2. r → 0 ⇒ c2 = constant ≡ 0.

• Hence, the solution of (11) becomes

φ(r) = 0. (13)
• In view of (13), (8) reduces to
~
~ = − ∂A .
E (14)
∂t
Home Page
• Substituting (8) and (14) into (4) and manipulating, we obtain
Title Page
1 ∂2A~
~−
∇2 A = 0, (15)
Contents
c2 ∂t2
JJ II
where we have used the identity ~a × (~b × ~c) = ~b(~a · ~c) − ~c(~a · ~b) and the
J I ~ = 0.
Lorentz gauge, ∇ · A
Page 6 of 37
• Equation (15) is similar to the ’wave equations’ for the electric and
Go Back magnetic fields! For a radiation field confined in a certain volume V ,
~ may be expressed as
the vector potential (A)
Full Screen

Close ~ r, t) = A
A(~ ~ (+) (~r, t) + A
~ (−) (~r, t), (16)
Quit
where
~ (+) (~r, t) = Cα(t)ei~k·~r û,
A (17)

~ (−) (~r, t) = Cα∗ (t)e−i~k·~r û,


A (18)
• Here, C is a real constant, ~k is the wave vector, û is a unit vector in
~ and α is some ’dynamical’ variable whose charac-
the direction of A,
teristics shall be determined later.

Home Page • Note that expression (16) along with (17) and (18) guarantees that
Title Page
the vector potential is real. Since and constitute a pair of complex
conjugate variables, each of these variables must satisfy eq. (15),
Contents
subject to the Coulomb gauge defined by (10). Hence, we may write
JJ II
~ (+)
1 ∂2A ~ (−)
1 ∂2A
~ (+) −
∇2 A = 0, ~ (−) −
∇2 A = 0, (19)
J I c2 ∂t2 c2 ∂t2
Page 7 of 37 and
~ (+) = 0,
∇·A ~ (−) = 0.
∇·A (20)
Go Back

Full Screen • Substituting (17) into (20), we find that (with ∇ → −ik)
Close ~ ~k · û = 0.
∇ · [Cα(t)eik·~r û] = 0, ⇒ (21)
Quit

• In addition, substituting (17) into (19), we get

21 d2 α d2 α
−k α − 2 2 = 0, 2
+ ω 2 α = 0, (22)
c dt dt
where ω = ck is the angular frequency.
• Equation (22) is a second-order ordinary differential equation. Its
general solution may be written as

α(t) = Aeiωt + Be−iωt ,


Home Page
where A and B are constants.
Title Page
• In our case (from physical consideration - α(t → ∞) must be finite),
Contents we set A = 0 and if α(t = 0) = α(0), then B = α(0). Therefore, the
JJ II solution of (22) becomes

J I α(t) = α(0)e−iωt , (23)

Page 8 of 37
• On account of (16), (17), (18), and (23), the vector potential becomes
Go Back
~ r, t) = C[α(0)ei(~k·~r−ωt) + α∗ (0)e−i(~k·~r−ωt) ]û,
A(~ (24)
Full Screen

~ = −(∂ A)/(∂t),
• Using (24) in (14), i.e., E ~ we get
Close

Quit
~ r, t) = iCω[α(0)ei(~k·~r−ωt) − α∗ (0)e−i(~k·~r−ωt) ]û,
E(~ (25)

• Moreover, from (5), the magnetic field can be expressed as


     
~ = ∇×A ~= ∂A z ∂A y ∂A x ∂A z ∂A y ∂A x
B − êx + − êy + − êz ,
∂y ∂z ∂z ∂x ∂x ∂y
(26)
where the êi ’s are the unit vectors along the x-, y-, and z-axes.

• Writing ~k = kx êx + ky êy + kz êz and û = ux êx + uy êy + uz êz (with u2 =


u2x + u2y + u2z = 1) in component forms, we find the components of
Home Page the magnetic field to be
Title Page ∂Az ∂Ay ~ ~
− = iC[α(0)ei(k·~r−ωt) − α∗ (0)e−i(k·~r−ωt) ](ky uz − kz uy ),
Contents
∂y ∂z
∂Ax ∂Az ~ ~
JJ II − = iC[α(0)ei(k·~r−ωt) − α∗ (0)e−i(k·~r−ωt) ](kz ux − kx uz ),
(27)
∂z ∂x
∂Ay ∂Ax ~ ~
J I − = iC[α(0)ei(k·~r−ωt) − α∗ (0)e−i(k·~r−ωt) ](kx uy − ky ux ),
∂x ∂y
Page 9 of 37

• Combining the components in (27), we obtain


Go Back

Full Screen
~ r, t) = iC[α(0)ei(~k·~r−ωt) − α∗ (0)e−i(~k·~r−ωt) ](~k × û),
B(~ (28)

Close

Quit
1.1. The Hamiltonian of the radiation field

• The energy of a radiation field confined in a volume V is given by


Z  
1 2 1 2
H= 0 E + B dν. (29)
2 ν µ0
• From (25) and (28), we have

E2 = E ~ ∗ = C 2 ω 2 [2α∗ α − α2 e2i(~k·~r−ωt) − α∗2 e−2i(~k·~r−ωt) ],


~ ·E (30)

using the identity, ~u · ~u = 1, and


Home Page
~ ∗ = C 2 [2α∗ α − α2 e2i(~k·~r−ωt) − α∗2 e−2i(~k·~r−ωt) ][(~k × ~u) · (~k × ~u)],
~ ·B
B2 = B
Title Page

But,
Contents

(~k × ~u) · (~k × ~u) = ~k · [~u × (~k × ~u)] = ~k · [~k(~u · ~u) − ~u(~u · ~k)] = k 2 ,
JJ II

J I
since ~u · ~k = 0 and ~u · ~u = 1. Then,
C 2ω2 ~ ~
Page 10 of 37 B = 2 [2α∗ α − α2 e2i(k·~r−ωt) − α∗2 e−2i(k·~r−ωt) ]k 2 ,
2
(31)
c
Go Back
• Substituting (30) and (31) into (29), we get
Z
Full Screen ~ ~
2 2
H = 0 C ω [2α∗ α − α2 e2i(k·~r−ωt) − α∗2 e−2i(k·~r−ωt) ]dν. (32)
Close
ν
p
where c2 = 1/(0 µ0 ) is used.
Quit

• Also, from the definition of the delta function, we have


Z
~
e±2ik·~r dν = π 3 δ(~k), (33)
ν

where δ(~k) is the Kronecker delta function.


• In view of (33), (32) takes the form

H = 20 C 2 ω 2 V α∗ α − 0 C 2 ω 2 π 3 [α2 e−2iωt + α∗2 e2iωt ]. (34)

Home Page • For a radiation field with ~k 6= 0, δ(~k) = 0. Therefore, the Hamiltonian
Title Page
of the radiation field reduces to

Contents H = 20 C 2 ω 2 V α∗ α. (35)


JJ II
• Let us choose the Hamiltonian of the radiation field to be
J I

Page 11 of 37
H = ~ωα∗ α. (36)

Go Back
• Comparing (35) and (36), we obtain
Full Screen r
~
C= . (37)
Close 20 ωV
Quit
• The quantum description of the radiation field is obtained by replac-
ing the dynamical variables α and α∗ by the operators â and ↠sub-
ject to the commutation relation

[â, ↠] = 1. (38)


• Hence, in the quantum theory of radiation, the vector potential, the
electric field, and the magnetic field are represented by
r
ˆ
~ r, t) = ~ ~ ~
A(~ [âei(k·~r−ωt) + ↠e−i(k·~r−ωt) ]~u, (39)
Home Page 20 ωV

Title Page r
~ˆ r, t) = i
E(~
~ω ~ ~
[âei(k·~r−ωt) − ↠e−i(k·~r−ωt) ]~u, (40)
Contents 20 V
JJ II and r
~ˆ r, t) = i
B(~
~ ~ ~
[âei(k·~r−ωt) − ↠e−i(k·~r−ωt) ](~k × ~u). (41)
J I 20 ωV
Page 12 of 37 • Equations (39) - (41) represents the quantized electromagnetic fields.
Go Back It is worth noting that the position vector ~r which appears in expres-
sions (39)-(41) is a c-number variable and not an operator.
Full Screen

Close • Moreover, when applying the above quantized fields to the classical
Hamiltonian it essential to determine the form of operators ordering
Quit
as a function of â and ↠.

• Essentially, there are three forms of operator ordering: normal, anti-


normal, and symmetric ordering.
1. In normal ordering, all the â operators are placed to the right of
the ↠operators.
2. In antinormal ordering, all the ↠operators are placed to the left
Home Page of the ↠operators.

Title Page 3. On the other hand, symmetric ordering involves the average of
all possible orderings of â and ↠.
Contents

JJ II • It happens that the quantum theory of radiation is compatible with


observations when the Hamiltonian is in the normal order. There-
J I
fore, the quantum Hamiltonian of a single-mode radiation takes the
Page 13 of 37
form
Go Back Ĥ = ~ω↠â = ~ωn̂. (42)
Full Screen where n̂ = ↠â, known as the number operator - the number of pho-
Close
tons in the radiation.

Quit
2. The Number State

• The eigenvalue equation of the number operator can be expressed


as
Home Page
n̂ | ni = n | ni. (43)
Title Page
where | ni is called the number state. In order to determine the ac-
Contents
tion of the operator â on the number state | ni, we may write
JJ II
â | ni = c | mi. (44)
J I
where its complex conjugate is
Page 14 of 37

Go Back hn | ↠= c∗ hm | . (45)

Full Screen
• Taking the inner product, we obtain
Close
hn | ↠â | ni = hn | n̂ | ni =| c |2 hm | mi ⇒ nhn | ni =| c |2 hm | mi.
Quit
(46)

• Assuming hn | ni and hm | mi to be normalized and taking c to be


real, we find that

c= n. (47)
• In view of (47), (44) can be rewritten as

â | ni = n | mi. (48)

Home Page • Furthermore, multiplying (48) by ↠, we have


Title Page √
↠â | ni = n↠| mi. (49)
Contents

and
JJ II √
hn | ↠â = nhm | â, (50)
J I
so that taking the inner product of (49) and (50), we get
Page 15 of 37

hn | ↠â↠â | ni = nhm | â↠| mi, (51)


Go Back

Full Screen or
Close
hm | â↠| mi = n (52)

Quit
since hn | ↠â↠â | ni = n2 .

• Then, applying the commutation relation, [â, ↠] = â↠− ↠â = 1, we


get

hm | â↠| mi = hm | ↠â + 1 | mi = hm | ↠â | mi + hm | mi = m + 1


• Comparison of this result (52) gives

m+1=n ⇒ m=n−1 (53)

Home Page • Hence, the action of the operator â on the number state | ni is
Title Page √
â | ni = n | n − 1i. (54)
Contents

JJ II • Similarly, it can be shown that the action of the operator ↠on the
number state | ni is
J I

Page 16 of 37 ↠| ni = n + 1 | n + 1i. (55)
Go Back
Remarks: The number state | ni represents the state of a single-
Full Screen mode light with a definite number of photons. Since the action of
Close â is to reduce the number of photons by one, it is called the annihi-
lation operator. On the other hand, the action of ↠is to increase the
Quit
number of photons by one, it is called the creation operator.

• Next, consider the equation

hn | n̂ | ni = hn | ↠â | ni. (56)


• Assuming that | ni is normalized, this can be written as

hφ | φi = n, (57)

where | φi = â | ni. Since hφ | φi ≥ 0, we find that n ≥ 0. Moreover,


Home Page
since n is the eigenvalue of a Hermitian operator, n must be real.
Title Page
Also note that
Contents h0 | n̂ | 0i = h0 | ↠â | 0i = 0, (58)
JJ II implies
J I â | 0i = 0, (59)

Page 17 of 37 • Furthermore, we see that


Go Back

↠| 0i = 1! | 1i,

Full Screen
â†2 | 0i = 2! | 2i,

Close â†3 | 0i = 3! | 3i,
Quit ... = ...,

â†n | 0i = n! | ni,

so that we obtain
â†n
| ni = √ | 0i, (60)
n!
where n = 0, 1, 2, ....

• Since the number states are eigenstates of a Hermitian operator, they


are orthogonal. Moreover, we consider the number states to be nor-
Home Page malized so that they become orthonormal, i.e.,
Title Page
hm | ni = δmn . (61)
Contents

JJ II • Therefore, in view of the fact that a set of orthonormal eigenstates is


complete, the number states must satisfy the completeness relation
J I

Page 18 of 37 X
Iˆ = | nihn |. (62)
Go Back n=0

Full Screen

Close

Quit
3. The Coherent State

• The displacement operator (D̂) is defined by


† −α∗ â)
Home Page
D̂(α) = e(αâ , (63)
Title Page
where α is a complex number.
Contents

• A coherent state | αi may be defined in terms of the displacement as


JJ II

J I
| αi = D̂(α) | 0i. (64)
Page 19 of 37
Since
Go Back D̂† (α) = D̂(−α),
Full Screen
• The displacement operator is unitary. That is, D̂(α)D̂† (α) = D̂† (α)D̂(α) =
Close ˆ where Iˆ is the identity operator.
I,
Quit
• Suppose f (↠, â) is a function that can be expanded in a power series.
Then, the following commutation relation is valid:


[â, f (↠, â)] = f (↠, â). (65)
∂â†
Thus,
∂ h † ∗
i
[â, D̂(α)] = exp(αâ − α â) ⇒ [â, D̂(α)] = αD̂(α). (66)
∂â†

Home Page • Expanding (66), we get


Title Page
âD̂(α) − D̂(α)â = αD̂(α).
Contents

JJ II
• Multiplying this by D̂(−α) from the left, we get

J I D̂(−α)âD̂(α) − D̂(−α)D̂(α)â = αD̂(−α)D̂(α) (67)


Page 20 of 37 D̂(−α)âD̂(α) = â + α.

Go Back
• We also have

Full Screen
[↠, f (↠, â)] = − f (↠, â). (68)
∂â
Close
Thus,
Quit ∂ h i
[↠, D̂(α)] = − exp(α↠−α∗ â) ⇒ [↠, D̂(α)] = α∗ D̂(α). (69)
∂â

• Expanding (69), we get

↠D̂(α) − D̂(α)↠= α∗ D̂(α).


and

D̂(−α)↠D̂(α) − D̂(−α)D̂(α)↠= α∗ D̂(−α)D̂(α)


D̂(−α)↠D̂(α) = ↠+ α∗ . (70)
Home Page

Title Page • Note that


hα | αi = h0 | D̂(−α)D̂(α) | 0i = h0 | 0i,
Contents
ˆ Also, since hn | ni = h0 | 0i = 1, we obtain
since D̂(α)D̂† (α) = I.
JJ II

J I hα | αi = 1. (71)

Page 21 of 37
This means that a coherent state | αi is normalized.
Go Back
• Consider the eigenvalue equation
Full Screen
â | αi = D̂(α)D̂(−α)âD̂(α) | 0i (72)
Close
= D̂(α)[â + α] | 0i
Quit
= αD̂(α) | 0i + D̂(α)â | 0i,

• Therefore, since D̂(α) | 0i =| αi and â | 0i = 0, we get

â | αi = α | αi. (73)
• The result obtained in (73) implies that the coherent state is an eigen-
state of the annihilation operator, â.

• Consider the Baker-Hausdorff identity


Home Page 1
e eB̂ = eÂ+B̂+ 2 [Â,B̂] , (74)
Title Page
provided that [Â, [Â, B̂]] = [B̂, [Â, B̂]] = 0. Using (74), D̂(α) can be
Contents
put in the normally-ordered form as:
JJ II † −α∗ â)
D̂(α) = e(αâ ,
J I

Page 22 of 37 |α|2 † ∗
D̂(α) = e− 2 eαâ e−α â , (75)
Go Back
since [α↠, α∗ â] = − | α |2 [↠, â] with [â, ↠] = 1. Note also that
Full Screen
∗ â
e−α | 0i = 0,
Close

Quit
since ∞
−α∗ â
X (−α∗ â)n α∗2 â2
e = = 1 − α∗ â + + ...
n=0
n! 2!
so that, a coherent state | αi can be expressed as
|α|2 †
| αi = e− 2 eαâ | 0i,
or,


|α|2 X αn
| αi = e 2 √ | ni, (76)
n=0 n!
since
Home Page â†n
| ni = √ | 0i.
Title Page
n!

Contents
• Furthermore, we have

|α|2 X αm
JJ II hn | αi = e− 2 √ δmn ,
m=0 m!
J I
or
Page 23 of 37 αn |α|2
hn | αi = e− √ , 2 (77)
Go Back
n!
where the relation hm | ni = δmn is used.
Full Screen

• Consider two coherent states hα | and | βi. That is,


Close


|α|2 X α∗n
Quit hα |= e 2 √ hn |, (78)
n=0 n!

and ∞

|β|2 X βm
| βi = e 2 √ | mi. (79)
m=0 m!
• The inner product of hα | and | βi is given by
1 2 2
X α∗n β m
hα | βi = e− 2 (|α| +|β| ) √ √ hn | mi.
n,m n! m!
1 2 +|β|2 )
X (α∗ β)n
Home Page
hα | βi = e− 2 (|α| ,
n
n!
Title Page
which may be rewritten as
Contents
∗ β− 1 |α|2 − 1 |β|2 )
hα | βi = e(α 2 2 . (80)
JJ II

J I • Note that

Page 24 of 37 | hα | βi |2 = hα | βihα | βi∗ (81)


∗ +α∗ β−|α|2 −|β|2 )

Go Back
= e(αβ
2
= e−|α−β| .
Full Screen

Close • Next, we will derive the completeness relation for coherent states.
Z XZ
2
Quit d α | αihα | = d2 α | nihn | αihα | mihm | (82)
n,m
X | nihm | Z 2
= √ d2 αe−|α| αn α∗m ,
n,m n!m!
where d2 α = d[<(α)]d[=(α)].
• Consider the integral in (82):
Z Z
−|α|2 n ∗m ∗ ∗
2
d αe α α = d2 αe(−αα +aα+bα ) αn α∗m |a=b=0 , (83)
∂n ∂m
Z
∗ ∗
Home Page = n m
d2 αe(−αα +aα+bα ) |a=b=0 ,
∂a ∂b
∂ n ∂ m ab
Z
Title Page ∗
= n m
e d2 αe−(a−α )(b−α) |a=b=0 ,
∂a ∂b
Contents
∂n ∂m
= π n m eab |a=b=0 .
JJ II ∂a ∂b
J I since Z Z ∞ Z ∞
−αα∗ −x2 2

Page 25 of 37
2
d αe = dxe dye−y = π
−∞ −∞

Go Back with a = b = 0 and α = x + iy. Further, we may rewrite (83) as


Full Screen
Z X 1 ∂n ∂m
2 −|α|2 n ∗m
d αe α α =π n m
(a` b` ) |a=b=0 . (84)
`
`! ∂a ∂b
Close

Quit
• Also, we have the identity

∂m n n! n−m ∂
m
x = x (a` b` ) |a=b=0 . (85)
∂xm (n − m)! ∂bm
• Using (85) and the condition that a = b = 0, we obtain
Z X1
2 `! `!
d2 αe−|α| αn α∗m = π δ`n δ`m , (86)
`
`! (` − n)! (` − m)!
Home Page
n!
= π δnm .
(n − m)!
Title Page

Contents
• In view of (86), (82) becomes
Z X | nihm | n!
JJ II d2 α | αihα | = π √ δnm , (87)
n,m n!m! (n − m)!
J I ∞
X
Page 26 of 37
= π | nihn |,
n=0
Go Back
which implies that
Full Screen ∞ Z
X 1
Close
Iˆ = | nihn | = d2 α | αihα |. (88)
n=0
π
Quit
• Equation (88) represents the completeness relation for coherent states.
Remarks: A single-mode light refers to light of definite energy (fre-
quency or wavelength). Coherent light is essentially a single-mode
light.
3.1. The mean photon number

• The mean photon number of a single-mode light is obtained as fol-


lows:
Home Page

Title Page
hn̂i = Tr(ρ̂n̂), (89)
X∞

Contents = Tr(ρ̂n̂ | nihn |),


n=0
JJ II X∞
= hn | ρ̂n̂ | ni,
J I n=0
X∞
Page 27 of 37
= nhn | ρ̂ | ni,
Go Back
n=0

X
Full Screen ⇒ hn̂i = nP (n),
n=0
Close
where P (n) = hn | ρ̂ | ni is known as the photon number distribution.
Quit
It represents the probability for the number of photons to be n.

• For a coherent state | βi, the density operator is defined by

ρ̂ =| βihβ | . (90)
• Thus, the photon number distribution for this state is

P (n) = hn | ρ̂ | ni = hn | βihβ | ni (91)


= hn | βihn | βi∗ =| hn | βi |2
n ∗n
2 β 2 β
Home Page 1 1
= e− 2 |β| √ e− 2 |β| √
Title Page n! n!
−|β|2
e
Contents = (ββ ∗ )n
n!
2
JJ II e−|β|
⇒ P (n) = | β |2n .
n!
J I

Page 28 of 37
• Also, we have

Go Back hn̂i = n = hβ | n̂ | βi (92)


Full Screen = hβ | ↠â | βi

Close
= β ∗ βhβ | βi

Quit
⇒ hn̂i = | β |2 .

• Consequently,
1 −n n
P (n) = e n . (93)
n!
The photon number distribution follows the Poisson’s distribution.
3.2. The variance of the photon number distribution

• The variance of the photon number distribution is defined by

Home Page (∆n)2 = hn̂2 i − hn̂i2 (94)


Title Page
• Note that
Contents

JJ II hn̂2 i = hβ | n̂n̂ | βi (95)


= hβ | ↠â↠â | βi
J I
= hβ | ↠(↠â + 1)â | βi
Page 29 of 37
= hβ | (â†2 â2 + ↠â) | βi
Go Back
⇒ hn̂2 i = β ∗2 β 2 + β ∗ β.
Full Screen

Since β ∗2 β 2 = (β ∗ β)2 = hn̂i2 , the variance becomes


Close

Quit (∆n)2 = β ∗ β = n. (96)

• It means that for a coherent light the variance of the photon number
distribution is equal to the mean photon number.
4. The Chaotic State

Thermal light is an example of a light mode in chaotic state. Thermal light


is generated by a source in thermal equilibrium at some temperature. The
Home Page
density operator ρ̂ for chaotic light may be obtained by using Lagrangian
Title Page
multipliers and by maximizing the entropy of the system as shown in the
Contents next section.
JJ II

J I 4.1. The density operator


Page 30 of 37
• The entropy S of a system represented by the density operator ρ̂ is
Go Back given
Full Screen
S = −Tr(ρ̂ ln ρ̂). (97)

Close • The change in the entropy δS corresponding to a small variation δ ρ̂


in the density operator is given by
Quit
h∂ i
δS = −Tr (ρ̂ ln ρ̂)δ ρ̂ . (98)
∂ ρ̂
• The entropy may be maximized by setting δS = 0. That is,
Tr[(1 + ln ρ̂)δ ρ̂] = 0. (99)
• If λ and β are Lagrangian multipliers, then

Tr(λρ̂) = λ, Tr(λδ ρ̂) = 0, (100)

Home Page and


hĤi = Tr(Ĥ ρ̂), Tr(β Ĥδ ρ̂) = 0. (101)
Title Page

Contents • Substituting (100) and (101) into (99), we obtain

JJ II
Tr[(1 + ln ρ̂ + λ + β Ĥ)δ ρ̂] = 0. (102)
J I

Page 31 of 37
• However, δ ρ̂ is arbitrary, so that (102) implies that

Go Back ln ρ̂ = −(1 + λ + β Ĥ). (103)


Full Screen
or
Close
ρ̂ = e−(1+λ) e−β Ĥ . (104)
Quit
• Applying the condition Tr(ρ̂) = 1, we obtain
1
e−(1+λ) = . (105)
Tr(e−β Ĥ )
• Therefore, eq. (104) becomes
e−β Ĥ
ρ̂ = . (106)
Tr(e−β Ĥ )
Home Page This is the required density operator for a chaotic light.
Title Page
• Employing the completeness relation and the equation Ĥ = ~ω↠â,
Contents we get
P∞ −β~ω↠â
JJ II e | nihn |
ρ̂ = P∞n=0 −β~ω↠â | mi
, (107)
J I m=0 hm | e
P −β~ωn
e
nP | nihn |
Page 32 of 37
= −β~ωm
.
me

Go Back Let
Full Screen
x = e−β~ω . (108)

Close
Then,
P∞ n
n=0 x | nihn |
Quit ρ̂ = P∞ m
, (109)
m=0 x
X∞
= (1 − x) xn | nihn |,
n=0
P∞
since m=0 xk = 1/(1 − x).
4.2. The mean photon number

• The mean photon number is defined by

Home Page n = Tr(ρ̂n̂), (110)


X∞
Title Page = Tr(ρ̂n̂ | nihn |),
n=0
Contents ∞
X
JJ II = (1 − x) nxn ,
n=0

J I d X n
= (1 − x)x nx ,
Page 33 of 37
dx n=0
d 1 
Go Back = (1 − x)x ,
dx 1 − x
x
Full Screen = ,
1−x
Close
or
Quit x e−β~ω 1
n= = = . (111)
1−x 1 − e−β~ω eβ~ω − 1
or
n 1
x= 1−x= . (112)
1+n 1+n
• Substituting (112) into (109), we get

X nn
ρ̂ = | nihn |. (113)
n=0
(1 + n)n+1
Home Page
This is the density operator for chaotic light.
Title Page

Contents
4.3. The variance and photon number distribution
JJ II
• The photon number distribution is determined by P (n) = hn | ρ̂ | ni,
J I
which results to
Page 34 of 37 nn
P (n) = . (114)
(1 + n)n+1
Go Back
This is the Bose-Einstein distribution function.
Full Screen

Close

Quit
• The variance of the photon number is obtained as follows:

hn̂2 i = Tr(ρ̂n̂2 ), (115)


X∞

Home Page
= Tr(ρ̂n̂2 | nihn |),
n=0

Title Page X
= (1 − x) n2 xn ,
Contents n=0
 d  d  X∞
JJ II = (1 − x) x x xn ,
dx dx n=0
J I  d  d  1 
= (1 − x) x x ,
Page 35 of 37 dx dx 1 − x
x  x 2
Go Back = +2 .
1−x 1−x
Full Screen
Since x = n/(1 + n), we obtain
Close
hn̂2 i = n + 2n2 , (116)
Quit

and the variance of the photon number takes the form:

(∆n)2 = n + n2 . (117)
4.4. The value of β

• Recall that the mean photon number is found to be


1
n= . (118)
Home Page
eβ~ω −1
Title Page • The average thermal energy of thermal light may be written as:
Contents ~ω 1
Ĥ = ~ωhn̂i = = , (119)
eβ~ω −1 β
JJ II
using the approximation exp (β~ω) ≈ 1 + β~ω.
J I
• According to the equipartition theorem, the average energy of a sys-
Page 36 of 37
tem for each degree of freedom is kT /2, where k is the Boltzmann’s
Go Back
constant and T is the absolute temperature. Thermal light has two
Full Screen degrees of freedom corresponding to the electric and magnetic fields.
Close
Therefore,
1
Quit Ĥ = kT = kT, or β = kT. (120)
β
• Consequently, the mean photon number takes the form:
1
n= . (121)
e~ω/kT − 1
Home Page

Title Page

Contents

JJ II

J I Thank you !
Page 37 of 37

Go Back

Full Screen

Close

Quit

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy