Dirac Matrices and Lorentz Spinors

1
2

representation of the Spin(3) rotation group is

x,

y , and z , which obey both commutation and

Background: In 3D, the spinor j =

constructed from the Pauli matrices
anticommutation relations
[ i , j ] = 2iijk k

and { i , j } = 2 ij 122 .

(1)

Consequently, the spin matrices

S = 4i =

1
2

(2)

commute with each other like angular momenta, [S i , S j ] = iijk S k , so they represent the
generators of the rotation group. Moreover, under finite rotations R(, n) represented by


M (R) = exp in S ,

(3)

the spin matrices transform into each other as components of a 3vector,

M 1 (R)S i M (R) = Rij S j .

(4)

In this note, I shall generalize this construction to the Dirac spinor representation of the
Lorentz symmetry Spin(3, 1).
Dirac Matrices generalize the anti-commutation properties of the Pauli matrices i to
3 + 1 Minkowski dimensions:
+ = 2g 144 .

(5)

The are 4 4 matrices, but there are several different conventions for their specific form
In my class I shall follow the same convention as the Peskin & Schroeder textbook, namely
the Weyl convention where in 2 2 block notations
!
0
1
22
0 =
,
~ =
122
0

+~

!
.

(6)

Note that the 0 matrix is hermitian while the 1 , 2 , and 3 matrices are anti-hermitian.
Apart from that, the specific forms of the matrices are not important, the Physics follows
from the anti-commutation relations (5).
1

Lorentz spin matrices generalize S = 4i rather than S = 12 . In 4D, the vector

product becomes the antisymmetric tensor product, so we define
def

S = S =

i
4 [ , ].

(7)

Thanks to the anti-commutation relations (5) for the matrices, the S obey the commutation relations of the Lorentz generators J = J . Moreover, the commutation relations
of the spin matrices S with the Dirac matrices are similar to the commutation relations
of the J with a Lorentz vector such as P .
Lemma:
[ , S ] = ig ig .

(8)

Proof: Combining the definition (7) of the spin matrices as commutators with the anticommutation relations (5), we have
=

1
2 { , }

1
2 [ , ]

= g 144 2iS .

(9)

Since the unit matrix commutes with everything, we have

[X, S ] =

i
2 [X, ]

for any matrix X,

(10)

and the commutator on the RHS may often be obtained from the Leibniz rules for the
commutators or anticommutators:
[A, BC] = [A, B]C + B[A, C] = {A, B}C B{A, C},
(11)
{A, BC} = [A, B]C + B{A, C} = {A, B}C B[A, C].
In particular,
[ , ] = { , } { , } = 2g 2g

(12)

and hence
[ , S ] =

i
2 [ , ]

Quod erat demonstrandum.

2

= ig ig .

(13)

Theorem: The S matrices commute with each other like Lorentz generators,
 
S ,S
= ig S ig S ig S + ig S .

(14)

Proof: Again, we use the Leibniz rule and eq. (9):

 




,S
= , S + , S




= ig ig + ig ig
= ig ig ig + ig




= ig g 2iS ig g + 2iS




ig g 2iS + ig g + 2iS

(15)

= 2g S 2g S 2g S + 2g S ,
and hence
 
S ,S
=

i
2

 
,S
= ig S ig S ig S + ig S .

(16)

Quod erat demonstrandum.

In light of this theorem, the S matrices represent the Lorentz generators J in a
4-component spinor multiplet.
Finite Lorentz transforms:
Any continuous Lorentz transform a rotation, or a boost, or a product of a boost and a
rotation obtains from exponentiating an infinitesimal symmetry
X 0 = X +  X

(17)

where the infinitesimal  matrix is antisymmetric when both indices are raised (or both
lowered),  =  . Thus, the L matrix of any continuous Lorentz transform is a matrix
exponential
L = exp() + +

1
2

1
6

(18)

of some matrix that becomes antisymmetric when both of its indices are raised or lowered,
= . Note however that in the matrix exponential (18), the first index of is raised
3

while the second index is lowered, so the antisymmetry condition becomes (g)> = (g)
instead of > = .
The Dirac spinor representation of the finite Lorentz transform (18) is the 4 4 matrix


MD (L) = exp 2i S .

(19)

The group law for such matrices

L1 , L2 SO+ (3, 1),

MD (L2 L1 ) = MD (L2 )MD (L1 )

(20)

follows automatically from the S satisfying the commutation relations (14) of the Lorentz
generators, so I am not going to prove it. Instead, let me show that when the Dirac matrices
are sandwiched between the MD (L) and its inverse, they transform into each other as
components of a Lorentz 4vector,
1
MD
(L) MD (L) = L .

(21)

This formula makes the Dirac equation transform covariantly under the Lorentz transforms.
Proof: In light of the exponential form (19) of the matrix MD (L) representing a finite Lorentz
transform in the Dirac spinor multiplet, lets use the multiple commutator formula (AKA
the Hadamard Lemma ): for any 2 matrices F and H,


exp(F )H exp(+F ) = H + H, F +

1
2


 
H, F , F +

1
6


  
H, F , F , F + . (22)

1
In particular, let H = while F = 2i S so that MD (L) = exp(+F ) and MD
(L) =

exp(F ). Consequently,


1
MD
(L) MD (L) = + , F +

1
2

  
,F ,F +

1
6

   
, F , F , F + (23)

where all the multiple commutators turn out to be linear combinations of the Dirac matrices.
4

Indeed, the single commutator here is



 
, F = 2i , S =

1
2

g g

= g = , (24)

while the multiple commutators follow by iterating this formula:

  


, F , F = , F = ,

   
, F , F , F = , . . . . (25)

Combining all these commutators as in eq. (23), we obtain



1
MD
MD = + , F +

1
2

  
,F ,F +

1
, F
6
1
6



  
,F ,F +

= + + 21 +
+


= + + 12 + 61 +

(26)

L .
Quod erat demonstrandum.

Dirac Equation and Dirac Spinor Fields

History:
Originally, the KleinGordon equation was thought to be the relativistic version of the
Schrodinger equation that is, an equation for the wave function (x, t) for one relativistic
particle. But pretty soon this interpretation run into trouble with bad probabilities (negative,
or > 1) when a particle travels through high potential barriers or deep potential wells. There
were also troubles with relativistic causality, and a few other things.
Paul Adrien Maurice Dirac had thought that the source of all those troubles was the
p
= p
2 + m2 in the coordinate basis, and that he
ugly form of relativistic Hamiltonian H
could solve all the problems with the Klein-Gordon equation by rewriting the Hamiltonian
as a first-order differential operator
= p

H
~ + m

Dirac equation i

= i~
+ m
t

(27)

where 1 , 2 , 3 , are matrices acting on a multi-component wave function. Specifically, all

5

four of these matrices are Hermitian, square to 1, and anticommute with each other,
{i , j } = 2ij ,

{i , } = 0,

2 = 1.

(28)

Consequently

2

~ p

= i j pi pj =

1
i pj
2 {i , j } p

2,
= ij pi pj = p

(29)

and therefore
2
H
Dirac =


2
+ m

~ p
=

2

~ p

+ {i , } pi m + 2 m2 = p2 + 0 + m2 . (30)

2 + m2 , as it should for the relativistic particle.

This, the Dirac Hamiltonian squares to p
The Dirac equation (27) turned out to be a much better description of a relativistic
electron (which has spin = 12 ) than the KleinGordon equation. However, it did not resolve
the troubles with relativistic causality or bad probabilities for electrons going through big
potential differences e > 2me c2 . Those problems are not solvable in the context of a
relativistic single-particle quantum mechanics but only in quantum field theory.
Modern point of view:
Today, we interpret the Dirac equation as the equation of motion for a Dirac spinor field
(x), comprising 4 complex component fields (x) arranged in a column vector

1 (x)

2 (x)

(x) =
(x) ,
3

(31)

4 (x)
and transforming under the continuous Lorentz symmetries x0 = L x according to
0 (x0 ) = MD (L)(x).

(32)

The classical EulerLagrange equation of motion for the spinor field is the Dirac equation
i

+ i~
m = 0.
t
6

(33)

To recast this equation in a Lorentz-covariant form, let

= 0,

i = 0 i ;

(34)

it is easy to see that if the matrices obey the anticommutation relations (5) then the
~
and matrices obey the relations (28) and vice verse. Now lets multiply the whole LHS of
the Dirac equation (33) by the = 0 :



0 = 0 i0 + i 0~ m 0 (x) = i 0 0 + i i i m)(x),

(35)

i m (x) = 0.

(36)

and hence

2
2 + m2 , the Dirac equation for the spinor field implies the
As expected from H
Dirac = p
KleinGordon equation for each component (x). Indeed, if (x) obey the Dirac equation,
then obviously




i m i m (x) = 0,

(37)

but the differential operator on the LHS is equal to the KleinGordon m2 + 2 times a unit
matrix:




i m i m = m2 + = m2 + 12 { , } = m2 + g .
(38)
The Dirac equation (36) transforms covariantly under the Lorentz symmetries
its LHS transforms exactly like the spinor field itself.
Proof: Note that since the Lorentz symmetries involve the x coordinates as well as the
spinor field components, the LHS of the Dirac equation becomes


i 0 m 0 (x0 )

(39)

where
0

=
=
0
0
x
x
x
7

L1

(40)

Consequently,
0 0 (x0 ) =

L1

0 0 (x0 ) =

L1

MD (L) (x)

(41)

MD (L) (x).

(42)

MD (L) = L MD (L)


L1 MD (L) = MD (L) ,

(43)

and hence

But according to eq. (23),

1
MD
(L) MD (L) = L

=
=

so
0 0 (x0 ) = MD (L) (x).

(44)

Altogether,


i m (x)
Lorentz

i 0 m 0 (x0 ) = MD (L) i m (x),

(45)

which proves the covariance of the Dirac equation. Quod erat demonstrandum.

Dirac Lagrangian
The Dirac equation is a first-order differential equation, so to obtain it as an Euler
Lagrange equation, we need a Lagrangian which is linear rather than quadratic in the spinor
fields derivatives. Thus, we want


L = i m

(46)

where (x) is some kind of a conjugate field to the (x). Since is a complex field, we
treat it as a linearly-independent from the , so the EulerLagrange equation for the
0 =

L
L

=

i m 0

immediately gives us the Dirac equation for the (x) field.

8

(47)

R
To keep the action S = d4 xL Lorentz-invariant, the Lagrangian (46) should transform
as a Lorentz scalar, L0 (x0 ) = L(x). In light of eq. (19) for the (x) field and covariance (45)
of the Dirac equation, the conjugate field (x) should transform according to
0

1
(x0 ) = (x) MD
(L)

L0 (x0 ) = L(x).

(48)

1
Note that the MD (L) matrix is generally not unitary, so the inverse matrix MD
(L) in

eq. (48) is different from the hermitian conjugate MD

(L). Consequently, the conjugate

field (x) cannot be identified with the hermitian conjugate field (x), since the latter
transforms to

1
0 (x0 ) = (x) MD
(L) 6= (x) MD
(L).

(49)

Instead of the hermitian conjugate, we are going to use the Dirac conjugate spinor, see below.
Dirac conjugates:
Let be a 4-component Dirac spinor and be any 4 4 matrix; we define their Dirac
conjugates according to

= 0 ,

= 0 0 .

(50)

Thanks to 0 0 = 1, the Dirac conjugates behave similarly to hermitian conjugates or

transposed matrices:
For a a product of 2 matrices, (1 2 ) = 2 1 .
Likewise, for a matrix and a spinor, ( ) = .
The Dirac conjugate of a complex number is its complex conjugate, (c 1) = c 1.


For any two spinors 1 and 2 and any matrix , 1 2 = 2 1 .
The Dirac spinor fields are fermionic, so they anticommute with each other, even


in the classical limit. Nevertheless, = + , and therefore for any


matrix , 1 2 = + 2 1 .
9

The point of the Dirac conjugation (50) is that it works similarly for all four Dirac
matrices ,
= + .

(51)

Proof: For = 0, the 0 is hermitian, hence

0 = 0 ( 0 ) 0 = + 0 0 0 = + 0 .

(52)

For = i = 1, 2, 3, the i are anti-hermitian and also anticommute with the 0 , hence
i = 0 ( i ) 0 = 0 i 0 = + 0 0 i = + i .

(53)

Corollary: The Dirac conjugate of the matrix

MD (L) = exp 2i S

(19)

representing any continuous Lorentz symmetry L = exp() is the inverse matrix

1
M D (L) = MD
(L) = exp + 2i S .

(54)

X = 2i S = + 18 [ , ] = + 41

(55)

Proof: Let

for some real antisymmetric Lorentz parameters = . The Dirac conjugate of the
X matrix is
X =

1

4

1
4

1
4

1
4

= 14 = X. (56)

Consequently,
X 2 = X X = +X 2 ,

X 3 = X X 2 = X 2 X = X 3 ,

...,

X n = (X)n ,
(57)

10

and hence
exp(X) =

X 1
X 1
Xn =
(X)n = exp(X).
n!
n!
n
n

(58)

In light of eq. (55), this means

exp 2i S

= exp + 2i S ,

(59)

that is,
1
M D (L) = MD
(L).

(60)

Quod erat demonstrandum.

Back to the Dirac Lagrangian:
Thanks to the theorem (60), the conjugate field (x) in the Lagrangian (46) is simply the
Dirac conjugate of the Dirac spinor field (x),

(x) = (x) 0 ,

(61)

which transforms under Lorentz symmetries as

0

1
(x0 ) = 0 (x0 ) = MD (L) (x) = (x) M D (x) = (x) MD
(L).

(62)

Consequently, the Dirac Lagrangian

L = i m = 0 i m

is a Lorentz scalar and the action is Lorentz invariant.

11

(46)

Hamiltonian for the Dirac Field

Canonical quantization of the Dirac spinor field (x) just like any other field begins
with classical Hamiltonian formalism. Lets start with the canonical conjugate fields,

L
=
(0 )

i 0

= i

(63)

the canonical conjugate to the Dirac spinor field (x) is simply its hermitian conjugate
(x). This is similar to what we had for the non-relativistic field, and it happens for the
same reason the Lagrangian which is linear in the time derivative.
In the non-relativistic field theory, the conjugacy relation (63) in the classical theory
lead to the equal-time commutation relations in the quantum theory,


t), (y,
t) = 0,
(x,



(x, t), (y, t) = 0,

However, the Dirac spinor field describes spin =

1
2


t), (y, t) = (3) (x y). (64)
(x,

particles like electrons, protons, or

neutrons which are fermions rather than bosons. So instead of the commutations relations (64), the spinor fields obey equal-time anti-commutation relations

(x, t),
(y, t) = 0,



(x, t),
(y, t) = 0,


(x, t),
(y, t) = (3) (x y).

(65)

Next, the classical Hamiltonian obtains as

Z
H =

d3 x H(x),

H = i 0 L

(66)
0

= i 0 i 0 + i~ m


= i 0~ + 0 m
where the terms involving the time derivative 0 cancel out. Consequently, the Hamiltonian
12

operator of the quantum field theory is

=
H

(x) i 0~ + 0 m (x).

d3 x

(67)

Note that the derivative operator (i 0~ + 0 m) in this formula is precisely the 1-particle
Dirac Hamiltonian (27). This is very similar to what we had for the quantum non-relativistic
fields,
=
H

d x (x)


1 2

+ V (x) (x),
2M

(68)

except for a different differential operator, Schrodinger instead of Dirac.

In the Heisenberg picture, the quantum Dirac field obeys the Dirac equation. To see
how this works, we start with the Heisenberg equation



i
(x, t) = (x, t), H =
t



(x, t), H(y,

t) ,
d3 y

(69)

and then evaluate the last commutator using the anti-commutation relations (65) and the
Leibniz rules (11). Indeed, lets use the Leibniz rule
[A, BC] = {A, B}C B{A, C}

(70)

for
(x, t),
A =
(y, t),
B =

C =

(71)

(y, t),
i 0~ + 0 m

so that BC = H(y,
t). For the A, B, C at hand,
{A, B} = (3) (x y)

(72)

while
{A, C} =

(x, t),
(y, t)} = (diff.op.) 0 = 0.
i 0~ y + 0 m {
13

(73)

Consequently


(x, t), H(y,

t) [A, BC]
= {A, B} C B {A, C}


(y, t)
= (3) (x y) i 0~ + 0 m

(74)

0,
hence


(x, t), H

Z
=
=

(y, t)
d3 y (3) (x y) i 0~ + 0 m


(x, t),
i 0~ + 0 m

(75)

and therefore

i0 (x,
t) =

t).
i 0~ + 0 m (x,

(76)

Or if you prefer,

= 0.
i m (x)

14

(77)