• 検索結果がありません。

The wave equations for the massless (m = 0) spin-1 field are the Maxwell equations, and for the massive (m > 0) spin-1 field the Proca equations.

N/A
N/A
Protected

Academic year: 2021

シェア "The wave equations for the massless (m = 0) spin-1 field are the Maxwell equations, and for the massive (m > 0) spin-1 field the Proca equations."

Copied!
5
0
0

読み込み中.... (全文を見る)

全文

(1)

3 Wave equations for particles with spin 1

The wave equations for the massless (m = 0) spin-1 field are the Maxwell equations, and for the massive (m > 0) spin-1 field the Proca equations.

3.1 Maxwell equations (in vacuum)

The first set of Maxwell equations for the electric and magnetic fields is

∇ × ~ E ~ + B ~ ˙ = 0

∇ · ~ B ~ = 0 (3.1)

The second set of Maxwell equations is

∇ × ~ B ~ − E ~ ˙ = 0

∇ · ~ E ~ = 0 (3.2)

The 4-vector potential A

µ

= φ, ~ A

is defined by the equations (see Sect. 10 or RQM1)

E ~ = − ∇φ ~ − A , ~ ˙ B ~ = ∇ × ~ A ~ (3.3) Then the first set of equations (3.1) is satisfied automatically!

In order to express the equations (3.2) in terms of the vector potential, we use the field strength tensor F

µν

defined by

F

µν

= ∂

µ

A

ν

− ∂

ν

A

µ

(3.4)

The components F

µν

are related to the electric and magnetic fields by (see Eq.(3.3)) F

i0

= −∇

i

A

0

− ∂

0

A

i

= E

i

, F

ij

= ∂

i

A

j

− ∂

j

A

i

= −

∇ × ~ A ~

k

= −B

k

[(i, j, k) is a cyclic permutation of (1, 2, 3).] Then the second set of Maxwell equations (3.2) can be expressed in the compact form

ν

F

νµ

= 0 ⇔ A

µ

− ∂

µ

(∂ · A) = 0 (3.5) because of: ∂

i

F

i0

= 0 ⇒ ∇ · ~ E ~ = 0 and ∂

0

F

0i

+ ∂

j

F

ji

= 0 ⇒ − E ˙

i

+ ∇

j

B

k

− ∇

k

B

j

= 0, which gives

∇ × ~ B ~ − E ~ ˙ = 0. One can choose a gauge (“Lorentz gauge”) where ∂ · A = 0, then (3.5) simplifies to

A

µ

= 0.

(2)

3.2 Massive spin-1 field equation (Proca equation)

If we add a “mass term” (similar to the Klein-Gordon equation) m

2

A

µ

to the Maxwell equation (3.5), we get the Proca equation

ν

F

νµ

+ m

2

A

µ

= 0

⇒ A

µ

− ∂

µ

(∂ · A) + m

2

A

µ

= 0 (3.6)

If we apply ∂

µ

to this equation, we get the relation ∂

µ

A

µ

= 0. For the massless case (Maxwell equation), this was only a choice of gauge, but for the massive case it must be satisfied. Therefore the Proca equations (3.6) are equivalent to the following set of equations:

+ m

2

A

µ

= 0 (3.7)

µ

A

µ

= 0 (3.8)

Relations like (3.8) are called constraints. Because of the constraint, there are three independent components (degrees of freedom) of the field A

µ

, namely 4 (components of A

µ

) - 1 (constraint) = 3 (degrees of freedom), as it should be for a massive spin-1 particle: The component of the spin vec- tor along the “spin quantization axis” (we will use the z-axis) has three possible values λ = −1, 0, +1.

Plane wave solutions of the Proca equation:

The solutions with definite momentum ~ p and spin component λ = −1, 0, +1 are

A

µ

(x) = ε

µ

(~ p, λ) e

−i(Et−~p·~x)

(3.9) Here E = ± p

~

p

2

+ m

2

= ±E

p

because of (3.7). Similar to the Klein-Gordon case, we call the solution with E = +E

p

the “positive frequency solution”, and the other with E = −E

p

the “negative frequency solution”. (We will show later that both solutions have positive energy.) ε

µ

(~ p, λ) is the spin part of the wave function, called the “polarization 4-vector”, which must satisfy the constraint (from Eq.(3.8))

p

µ

ε

µ

(~ p, λ) = 0 (3.10)

(i) In the rest frame of the particle p

µ

= (m, ~ 0), the polarization vector has the form

ε

µ

(~ p = 0, λ) = (0, ~ ) (3.11)

(3)

Here the set of three vectors ( ~

−1

, ~

0

, ~

1

) is not determined by the Proca equation, but can be chosen as eigenvectors of the z-component of the spin operator ( ˆ S

3

) with eigenvalues −1, 0, +1. Here we use the following spin matrices S ~ ˆ =

S ˆ

1

, S ˆ

2

, S ˆ

3

for a spin-1 particle (“adjoint representation”):

S ˆ

i

jk

= −i

ijk

, which satisfy the commutation relations h S ˆ

i

, S ˆ

j

i

= i

ijk

S ˆ

k

. The explicit forms are

S ˆ

1

=

0 0 0 0 0 −i 0 i 0

 , S ˆ

2

=

0 0 i 0 0 0

−i 0 0

 , S ˆ

3

=

0 −i 0 i 0 0

0 0 0

 (3.12)

The eigenvectors of ˆ S

3

with eigenvalues λ = −1, 0, +1 are then obtained as

~

−1

= 1

√ 2 (1, −i, 0) , ~

0

= (0, 0, 1) , ~

+1

= −1

√ 2 (1, i, 0) (3.13)

They satisfy the orthogonality and completeness relations

~

λ0

· ~

λ

= δ

λ0λ

, X

λ

iλ

j†λ

= δ

ij

(3.14)

(ii) In the frame where the particle has momentum ~ p, we must apply a Lorentz transformation with velocity ~ v = −~ p/E

p

to the 4-vectors (3.11):

ε

µ

(~ p, λ) = Λ

µν

(~ v) ε

ν

(~ p = 0, λ) =

p ~ · ~

λ

m , ~

λ

+ ~ p (~ p · ~

λ

) m (E

p

+ m)

(3.15) By construction, they satisfy the following relations (see (3.10) and (3.14)):

p

µ

ε

µ

(~ p, s) = 0 , ε

µ∗

(p, λ

0

) ε

µ

(p, λ) = −δ

λλ0

, ε

µ

(~ p, λ) = (−1)

λ

ε

µ

(~ p, −λ)

(3.16) X

λ

ε

µ∗

(p, λ) ε

ν

(p, λ) = −g

µν

+ p

µ

p

ν

m

2

(3.17)

3.3 Lagrangian and Hamiltonian for the Proca equation

The Proca equations (3.7), (3.8) follow from the following Lagrangian density:

L = − 1

4 F

µν

F

µν

+ m

2

2 A

µ

A

µ

= − 1

2 (∂

µ

A

ν

) (∂

µ

A

ν

) + 1

2 (∂

µ

A

ν

) (∂

ν

A

µ

) + m

2

2 A

2

(3.18)

(4)

Check this:

∂L

∂ (∂

µ

A

ν

) = −∂

µ

A

ν

+ ∂

ν

A

µ

= −F

µν

∂L

∂A

ν

= m

2

A

ν

and therefore the Euler-Lagrange equation for A

ν

µ

∂ L

∂ (∂

µ

A

ν

) = ∂L

∂A

ν

becomes the Proca equation (3.6).

Using the definition of the field strength tensor (see Eq.(3.4)), the Lagrangian density (3.18) can be expressed as

L = 1 2

E ~

2

− B ~

2

+ m

2

2

A

20

− A ~

2

(3.19) where (see Eq.(3.3))

E ~ = − ∇A ~

0

− A , ~ ˙ B ~ = ∇ × ~ A ~

The momenta conjugate to A

0

and A ~ are then obtained as Π

0

≡ ∂L

∂ A ˙

0

= 0 , Π ~ ≡ ∂L

∂ A ~ ˙

= − E ~ (3.20)

Then the Hamiltonian density becomes

H = Π

0

A ˙

0

+ Π ~ · A ~ ˙ − L = − E ~ · A ~ ˙ − L

=

E ~ + ∇A ~

0

· E ~ − 1 2

E ~

2

− B ~

2

− m

2

2

A

20

− A ~

2

= 1

2

E ~

2

+ B ~

2

+ m

2

A ~

2

− m

2

2 A

20

+ E ~ · ∇A ~

0

(3.21) The field A

0

can be eliminated by using the Proca field equation (first equation in 3.6) for µ = 0:

i

F

i0

+ m

2

A

0

= 0 ⇒ ∇ · ~ E ~ = −m

2

A

0

(3.22) Then the last term in (3.21) can be written in the form

E ~ · ∇A ~ = ∇ · ~ E A ~

− A

∇ · ~ E ~

= ∇ · ~ E A ~

+ m

2

A

2

(5)

Finally, the Hamiltonian (3.21) becomes

1

H =

Z

d

3

x H = 1 2

Z

d

3

x h

E ~

2

+ B ~

2

+ m

2

A

20

+ A ~

2

i

(3.23) This is positive definite, and therefore there are no negative energies for the Proca field. The inde- pendent (dynamical) fields are A ~ and E ~ , while A

0

and B ~ should be expressed as

A

0

= − ∇ · ~ E ~

m

2

, B ~ = ∇ × ~ A ~

In quantum field theory, the fields A ~ and E ~ become the dynamical quantum fields.

1The total derivative∇ ·~ E A~ 0

gives a surface term which vanishes after integration.

参照

関連したドキュメント

It should be noted that all these graphs are planar, even though it is more convenient to draw them in such a way that the (curved) extra arcs cross the other (straight) edges...

It is suggested by our method that most of the quadratic algebras for all St¨ ackel equivalence classes of 3D second order quantum superintegrable systems on conformally flat

[18] , On nontrivial solutions of some homogeneous boundary value problems for the multidi- mensional hyperbolic Euler-Poisson-Darboux equation in an unbounded domain,

By the algorithm in [1] for drawing framed link descriptions of branched covers of Seifert surfaces, a half circle should be drawn in each 1–handle, and then these eight half

Since the boundary integral equation is Fredholm, the solvability theorem follows from the uniqueness theorem, which is ensured for the Neumann problem in the case of the

Next, we prove bounds for the dimensions of p-adic MLV-spaces in Section 3, assuming results in Section 4, and make a conjecture about a special element in the motivic Galois group

We will give a different proof of a slightly weaker result, and then prove Theorem 7.3 below, which sharpens both results considerably; in both cases f denotes the canonical

The construction of homogeneous statistical solutions in [VF1], [VF2] is based on Galerkin approximations of measures that are supported by divergence free periodic vector fields