Si ce bandeau n'est plus pertinent, retirez-le. Cliquez ici pour en savoir plus.
Si ce bandeau n'est plus pertinent, retirez-le. Cliquez ici pour en savoir plus.
La mise en forme de cet article est à améliorer (août 2023 ).
La mise en forme du texte ne suit pas les recommandations de Wikipédia : il faut le « wikifier ».
Les matrices gamma forment des ensembles de matrices conventionnelles respectant des relations de commutations spécifiques.
En deux dimensions et avec la métrique euclidienne, cet ensemble de matrices s'identifie aux matrices de Pauli .
Elles sont définies comme l'ensemble des matrices complexes de dimensions 2 × 2 suivantes :
σ
1
=
(
0
1
1
0
)
{\displaystyle \sigma _{1}={\begin{pmatrix}0&1\\1&0\end{pmatrix}}}
σ
2
=
(
0
−
i
i
0
)
{\displaystyle \sigma _{2}={\begin{pmatrix}0&-\mathrm {i} \\\mathrm {i} &0\end{pmatrix}}}
σ
3
=
(
1
0
0
−
1
)
{\displaystyle \sigma _{3}={\begin{pmatrix}1&0\\0&-1\end{pmatrix}}}
(où i est l’unité imaginaire des nombres complexes). Les matrices de Pauli sont les génératrices du groupe SU(2).
Au sens large, il existe une 4e matrice gamma si on leur adjoint la matrice identité :
σ
0
=
(
1
0
0
1
)
.
{\displaystyle \sigma _{0}={\begin{pmatrix}1&0\\0&1\end{pmatrix}}\,.}
L'origine de ces matrices remonte aux tentatives de linéarisation par Dirac de l'équation de Klein-Gordon . Les matrices gamma sont en fait partie intégrante de l'équation de Dirac .
En notations contravariantes (voir Vecteur contravariant, covariant et covecteur ), les matrices de Dirac, de dimension 4, sont constituées par l'ensemble
{
γ
0
,
γ
1
,
γ
2
,
γ
3
}
{\displaystyle \{\gamma ^{0},\gamma ^{1},\gamma ^{2},\gamma ^{3}\}}
. Lorsqu'on parle de matrices gamma sans autre précision, on fait référence à ces matrices de Dirac.
D'une manière condensée on écrit cet ensemble comme suit :
{
γ
μ
}
{\displaystyle \{\gamma ^{\mu }\}}
avec
μ
=
0
,
1
,
2
,
3.
{\displaystyle \mu =0,1,2,3.}
En notations contravariantes, on définit par commodité une matrice
γ
5
{\displaystyle \gamma ^{5}}
en prenant le produit imaginaire des quatre matrices de Dirac comme suit :
γ
5
:=
i
γ
0
γ
1
γ
2
γ
3
=
−
i
4
!
ε
μ
ν
ρ
σ
γ
μ
γ
ν
γ
ρ
γ
σ
{\displaystyle \gamma ^{5}:=\mathrm {i} \gamma ^{0}\gamma ^{1}\gamma ^{2}\gamma ^{3}=-{\frac {\mathrm {i} }{4!}}\varepsilon _{\mu \nu \rho \sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }}
.
Bien que
γ
5
{\displaystyle \gamma ^{5}}
utilise la lettre gamma, ce n'est pas l'une des matrices gamma au sens strict. Le numéro 5 lui a été attribué du temps où l'ancienne notation de
γ
0
{\displaystyle \gamma ^{0}}
était "
γ
4
{\displaystyle \gamma ^{4}}
".
Utilisation du symbole de Levi-Civita
En dimension 4, le symbole de Levi-Civita est défini conventionnellement comme suit:
ε
i
j
k
l
=
{
+
1
si
(
i
,
j
,
k
,
l
)
est une permutation ''paire'' de
(
0
,
1
,
2
,
3
)
−
1
si
(
i
,
j
,
k
,
l
)
est une permutation ''impaire'' de
(
0
,
1
,
2
,
3
)
0
autrement .
{\displaystyle \varepsilon _{ijkl}={\begin{cases}+1&{\text{si }}(i,j,k,l){\text{ est une permutation ''paire'' de }}(0,1,2,3)\\-1&{\text{si }}(i,j,k,l){\text{ est une permutation ''impaire'' de }}(0,1,2,3)\\0&{\text{autrement .}}\end{cases}}}
Donc en particulier :
ε
0123
=
1
.
{\displaystyle \varepsilon _{0123}=1\,.}
On peut maintenant se poser la question de savoir ce que vaut ce même symbole en notations contravariantes :
ε
μ
ν
γ
δ
=
g
μ
α
g
ν
β
g
ρ
γ
g
σ
δ
ε
α
β
γ
δ
{\displaystyle \varepsilon ^{\mu \nu \gamma \delta }=g^{\mu \alpha }g^{\nu \beta }g^{\rho \gamma }g^{\sigma \delta }\varepsilon _{\alpha \beta \gamma \delta }}
Le seul produit utile des quatre composantes métriques étant nécessairement égal à -1 [= (+1)(-1)(-1)(-1)]
il vient :
ε
0123
=
−
1
.
{\displaystyle \varepsilon ^{0123}=-1\,.}
Notons aussi que dans ce contexte :
ε
i
j
k
l
=
{
+
1
si
(
i
,
j
,
k
,
l
)
est une permutation ''impaire'' de
(
0
,
1
,
2
,
3
)
−
1
si
(
i
,
j
,
k
,
l
)
est une permutation ''paire'' de
(
0
,
1
,
2
,
3
)
0
autrement .
{\displaystyle \varepsilon ^{ijkl}={\begin{cases}+1&{\text{si }}(i,j,k,l){\text{ est une permutation ''impaire'' de }}(0,1,2,3)\\-1&{\text{si }}(i,j,k,l){\text{ est une permutation ''paire'' de }}(0,1,2,3)\\0&{\text{autrement .}}\end{cases}}}
En dimension n , lorsque tous les i 1 ,...,in , j 1 ,...,j n prennent des valeurs 1, 2,..., n :
on a
ε
i
1
…
i
n
ε
j
1
…
j
n
=
n
!
δ
[
i
1
j
1
…
δ
i
n
]
j
n
=
δ
i
1
…
i
n
j
1
…
j
n
{\displaystyle \varepsilon _{i_{1}\dots i_{n}}\varepsilon ^{j_{1}\dots j_{n}}=n!\delta _{[i_{1}}{}^{j_{1}}\dots \delta _{i_{n}]}{}^{j_{n}}=\delta _{i_{1}\dots i_{n}}^{j_{1}\dots j_{n}}}
où
δ
i
1
…
i
n
j
1
…
j
n
{\displaystyle \delta _{i_{1}\dots i_{n}}^{j_{1}\dots j_{n}}}
est un delta généralisé de Kronecker de type (n,n).
Remarque 1 : Les crochets qui encadrent et figent une combinaison des n indices inférieurs signifient qu'il y a anticommutation complète entre ces n indices de sorte qu'on dénombre n! combinaisons de pareilles permutations. Au cas où certains indices seraient exclus de la commutation ces indices exclus seraient eux-mêmes encadrés par des barres verticales
Exemple :
T
α
[
β
|
γ
|
δ
]
{\displaystyle T_{\alpha [\beta |\gamma |\delta ]}}
est un tenseur covariant d'ordre 4 antisymétrique sur ces deuxième et quatrième indices seulement.
Remarque 2 :
ε
a
b
…
e
ε
f
b
…
e
=
(
n
−
1
)
!
δ
a
f
{\displaystyle \varepsilon _{ab\dots e}\varepsilon ^{fb\dots e}=(n-1)!\delta _{a}^{f}}
En dimension 4 il en découle que :
δ
μ
ν
ϱ
σ
α
β
γ
δ
=
i
4
!
ε
α
β
γ
δ
ε
μ
ν
ϱ
σ
{\displaystyle \delta _{\mu \nu \varrho \sigma }^{\alpha \beta \gamma \delta }={\frac {\mathrm {i} }{4!}}\varepsilon ^{\alpha \beta \gamma \delta }\varepsilon _{\mu \nu \varrho \sigma }}
.
En exploitant le fait que les quatre matrices gamma anticommutent, on obtient :
γ
0
γ
1
γ
2
γ
3
=
γ
[
0
γ
1
γ
2
γ
3
]
=
1
4
!
δ
μ
ν
ϱ
σ
0123
γ
μ
γ
ν
γ
ϱ
γ
σ
.
{\displaystyle \gamma ^{0}\gamma ^{1}\gamma ^{2}\gamma ^{3}=\gamma ^{[0}\gamma ^{1}\gamma ^{2}\gamma ^{3]}={\frac {1}{4!}}\delta _{\mu \nu \varrho \sigma }^{0123}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\varrho }\gamma ^{\sigma }.}
Et finalement il vient :
γ
5
=
i
γ
0
γ
1
γ
2
γ
3
=
i
4
!
ε
0123
ε
μ
ν
ϱ
σ
γ
μ
γ
ν
γ
ϱ
γ
σ
=
−
i
4
!
ε
μ
ν
ϱ
σ
γ
μ
γ
ν
γ
ϱ
γ
σ
.
{\displaystyle \gamma ^{5}=\mathrm {i} \gamma ^{0}\gamma ^{1}\gamma ^{2}\gamma ^{3}={\frac {\mathrm {i} }{4!}}\varepsilon ^{0123}\varepsilon _{\mu \nu \varrho \sigma }\,\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\varrho }\gamma ^{\sigma }=-{\frac {\mathrm {i} }{4!}}\varepsilon _{\mu \nu \varrho \sigma }\,\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\varrho }\gamma ^{\sigma }.}
Ceci est vrai dans l'hypothèse où la convention de sommation d'Einstein joue (ce qui justifie la division par 4!).
Dans l'hypothèse explicitement rappelée dans le texte où la convention de sommation ne jouerait pas, le dénominateur vaudrait 1 au lieu de 4!.
Intérêt de travailler avec des matrices gamma [ modifier | modifier le code ]
En principe, une matrice complexe 4x4 contient 16 éléments ayant chacun une partie réelle et une partie imaginaire, donc 32 paramètres au total. On pourrait donc penser que toute matrice complexe 4x4 est une combinaison linéaire de 32 matrices indépendantes.
Toutefois la propriété d'herméticité exigée de ces matrices réduisent ces 32 paramètres à 16 combinaisons bilinéaires
Γ
{\displaystyle \Gamma }
indépendantes permettant de former des scalaires de Lorentz de la forme
⟨
ψ
¯
Γ
ψ
⟩
{\displaystyle \langle {\overline {\psi }}\Gamma \psi \rangle }
et de construire des lagrangiens mettant en jeu des fermions.
Voici deux manières équivalentes de présenter ces 16 éléments :
première schématisation (dite de Pauli)
I
γ
1
γ
2
γ
3
γ
0
i
γ
2
γ
3
i
γ
3
γ
1
i
γ
1
γ
2
i
γ
1
γ
0
i
γ
2
γ
0
i
γ
3
γ
0
i
γ
1
γ
2
γ
3
i
γ
1
γ
2
γ
0
i
γ
3
γ
1
γ
0
i
γ
2
γ
3
γ
0
γ
1
γ
2
γ
3
γ
0
{\displaystyle {\begin{matrix}&&&&&I&&&&&\\&&\gamma ^{1}&&\gamma ^{2}&&\gamma ^{3}&&\gamma ^{0}&&\\\mathrm {i} \gamma ^{2}\gamma ^{3}&&\mathrm {i} \gamma ^{3}\gamma ^{1}&&\mathrm {i} \gamma ^{1}\gamma ^{2}&&\mathrm {i} \gamma ^{1}\gamma ^{0}&&\mathrm {i} \gamma ^{2}\gamma ^{0}&&\mathrm {i} \gamma ^{3}\gamma ^{0}\\&&\mathrm {i} \gamma ^{1}\gamma ^{2}\gamma ^{3}&&\mathrm {i} \gamma ^{1}\gamma ^{2}\gamma ^{0}&&\mathrm {i} \gamma ^{3}\gamma ^{1}\gamma ^{0}&&\mathrm {i} \gamma ^{2}\gamma ^{3}\gamma ^{0}\\&&&&&\gamma ^{1}\gamma ^{2}\gamma ^{3}\gamma ^{0}&&&&&\\\end{matrix}}}
Les différentes lignes de ce tableau peuvent être désignées comme suit : I,
γ
μ
,
γ
[
μ
ν
]
,
γ
[
μ
ν
ρ
]
e
t
γ
5
.
{\displaystyle \gamma ^{\mu },\gamma ^{[\mu \nu ]},\gamma ^{[\mu \nu \rho ]}\quad et\quad \gamma ^{5}.}
(voir le point "utilisation du symbole de Levi-Civita" pour plus de détails sur les crochets d'antisymétrisation)
Identité remarquable liée à ce schéma (Lemme 1 de Pauli)
I
γ
1
γ
2
γ
3
γ
0
S
01
S
02
S
03
S
12
S
13
S
23
γ
0
γ
5
γ
1
γ
5
γ
2
γ
5
γ
3
γ
5
γ
0
γ
1
γ
2
γ
3
{\displaystyle {\begin{matrix}&&&&&I&&&&&\\&&\gamma ^{1}&&\gamma ^{2}&&\gamma ^{3}&&\gamma ^{0}&&\\S^{01}&&S^{02}&&S^{03}&&S^{12}&&S^{13}&&S^{23}\\&&\gamma ^{0}\gamma ^{5}&&\gamma ^{1}\gamma ^{5}&&\gamma ^{2}\gamma ^{5}&&\gamma ^{3}\gamma ^{5}\\&&&&&\gamma ^{0}\gamma ^{1}\gamma ^{2}\gamma ^{3}&&&&&\\\end{matrix}}}
On peut démontrer l'indépendance des éléments I,
γ
μ
,
S
μ
ν
,
γ
μ
γ
5
e
t
γ
5
{\displaystyle \gamma ^{\mu },S^{\mu \nu },\gamma ^{\mu }\gamma ^{5}et\gamma ^{5}}
[ 1] .
Les matrices
S
i
j
{\displaystyle \quad S^{ij}}
sont définies supra.
Les matrices gamma font l'objet de plusieurs représentations. La plus immédiate est la représentation de Paul Dirac (appelée aussi la « représentation standard »).
Par la suite d'autres représentations ont été élaborées à partir de celle de Dirac.
Ainsi celle de Hermann Weyl s'obtient en effectuant une transformation unitaire à partir de celle de Dirac.
γ
Weyl
μ
=
U
γ
Dirac
μ
U
−
1
avec
U
=
1
2
(
1
1
−
1
1
)
,
U
−
1
=
U
†
=
1
2
(
1
−
1
1
1
)
{\displaystyle \gamma _{\text{Weyl}}^{\mu }=U\,\gamma _{\text{Dirac}}^{\mu }U^{-1}{\text{ avec }}U={\frac {1}{\sqrt {2}}}{\begin{pmatrix}1&1\\-1&1\end{pmatrix}},\ U^{-1}=U^{\dagger }={\frac {1}{\sqrt {2}}}{\begin{pmatrix}1&-1\\1&1\end{pmatrix}}\,}
, où
1
et
−
1
{\displaystyle 1{\text{ et }}-1}
sont des matrices identité (2x2).
Quant à la représentation de Ettore Majorana , elle est obtenue à partir de la « représentation standard » à l'aide de la matrice unitaire
U
{\displaystyle U}
suivante :
U
=
1
2
(
γ
D
0
γ
D
2
+
γ
D
0
)
.
{\displaystyle U={\frac {1}{\sqrt {2}}}(\gamma _{D}^{0}\gamma _{D}^{2}+\gamma _{D}^{0})\,.}
Construire la représentation de Majorana à partir de celle de Dirac
Cette représentation de Majorana a la propriété intéressante que toutes les matrices
γ
μ
{\displaystyle \gamma ^{\mu }}
sont imaginaires pures, ce qui rend les calculs commodes quand on considère les opérateurs de conjugaison de charge et de parité.
Représentation de
Dirac (D )
Weyl (W )
Majorana (M )
γ
D
0
=
(
1
0
0
0
0
1
0
0
0
0
−
1
0
0
0
0
−
1
)
{\displaystyle \gamma _{D}^{0}={\begin{pmatrix}1&0&0&0\\0&1&0&0\\0&0&-1&0\\0&0&0&-1\end{pmatrix}}}
γ
W
0
=
(
0
0
1
0
0
0
0
1
1
0
0
0
0
1
0
0
)
{\displaystyle \gamma _{W}^{0}={\begin{pmatrix}0&0&1&0\\0&0&0&1\\1&0&0&0\\0&1&0&0\end{pmatrix}}}
γ
M
0
=
(
0
0
0
−
i
0
0
i
0
0
−
i
0
0
i
0
0
0
)
{\displaystyle \gamma _{M}^{0}={\begin{pmatrix}0&0&0&-\mathrm {i} \\0&0&\mathrm {i} &0\\0&-\mathrm {i} &0&0\\\mathrm {i} &0&0&0\end{pmatrix}}}
γ
D
1
=
(
0
0
0
1
0
0
1
0
0
−
1
0
0
−
1
0
0
0
)
{\displaystyle \gamma _{D}^{1}={\begin{pmatrix}0&0&0&1\\0&0&1&0\\0&-1&0&0\\-1&0&0&0\end{pmatrix}}}
γ
W
1
=
(
0
0
0
1
0
0
1
0
0
−
1
0
0
−
1
0
0
0
)
{\displaystyle \gamma _{W}^{1}={\begin{pmatrix}0&0&0&1\\0&0&1&0\\0&-1&0&0\\-1&0&0&0\end{pmatrix}}}
γ
M
1
=
(
i
0
0
0
0
−
i
0
0
0
0
i
0
0
0
0
−
i
)
{\displaystyle \gamma _{M}^{1}={\begin{pmatrix}\mathrm {i} &0&0&0\\0&-\mathrm {i} &0&0\\0&0&\mathrm {i} &0\\0&0&0&-\mathrm {i} \end{pmatrix}}}
γ
D
2
=
(
0
0
0
−
i
0
0
i
0
0
i
0
0
−
i
0
0
0
)
{\displaystyle \gamma _{D}^{2}={\begin{pmatrix}0&0&0&-\mathrm {i} \\0&0&\mathrm {i} &0\\0&\mathrm {i} &0&0\\-\mathrm {i} &0&0&0\end{pmatrix}}}
γ
W
2
=
(
0
0
0
−
i
0
0
i
0
0
i
0
0
−
i
0
0
0
)
{\displaystyle \gamma _{W}^{2}={\begin{pmatrix}0&0&0&-\mathrm {i} \\0&0&\mathrm {i} &0\\0&\mathrm {i} &0&0\\-\mathrm {i} &0&0&0\end{pmatrix}}}
γ
M
2
=
(
0
0
0
i
0
0
−
i
0
0
−
i
0
0
i
0
0
0
)
{\displaystyle \gamma _{M}^{2}={\begin{pmatrix}0&0&0&\mathrm {i} \\0&0&-\mathrm {i} &0\\0&-\mathrm {i} &0&0\\\mathrm {i} &0&0&0\end{pmatrix}}}
γ
D
3
=
(
0
0
1
0
0
0
0
−
1
−
1
0
0
0
0
1
0
0
)
{\displaystyle \gamma _{D}^{3}={\begin{pmatrix}0&0&1&0\\0&0&0&-1\\-1&0&0&0\\0&1&0&0\end{pmatrix}}}
γ
W
3
=
(
0
0
1
0
0
0
0
−
1
−
1
0
0
0
0
1
0
0
)
{\displaystyle \gamma _{W}^{3}={\begin{pmatrix}0&0&1&0\\0&0&0&-1\\-1&0&0&0\\0&1&0&0\end{pmatrix}}}
γ
M
3
=
(
0
−
i
0
0
−
i
0
0
0
0
0
0
−
i
0
0
−
i
0
)
{\displaystyle \gamma _{M}^{3}={\begin{pmatrix}0&-\mathrm {i} &0&0\\-\mathrm {i} &0&0&0\\0&0&0&-\mathrm {i} \\0&0&-\mathrm {i} &0\end{pmatrix}}}
γ
D
5
=
(
0
0
1
0
0
0
0
1
1
0
0
0
0
1
0
0
)
{\displaystyle \gamma _{D}^{5}={\begin{pmatrix}0&0&1&0\\0&0&0&1\\1&0&0&0\\0&1&0&0\end{pmatrix}}}
γ
W
5
=
(
−
1
0
0
0
0
−
1
0
0
0
0
1
0
0
0
0
1
)
{\displaystyle \gamma _{W}^{5}={\begin{pmatrix}-1&0&0&0\\0&-1&0&0\\0&0&1&0\\0&0&0&1\end{pmatrix}}}
γ
M
5
=
(
0
−
i
0
0
i
0
0
0
0
0
0
i
0
0
−
i
0
)
{\displaystyle \gamma _{M}^{5}={\begin{pmatrix}0&-\mathrm {i} &0&0\\\mathrm {i} &0&0&0\\0&0&0&\mathrm {i} \\0&0&-\mathrm {i} &0\end{pmatrix}}}
On peut facilement voir à l'examen du tableau comparatif ci-dessus que la représentation de Weyl n'est autre que celle de Dirac où l'on a permuté (anticommuté) les matrices
γ
0
{\displaystyle \gamma ^{0}}
et
γ
5
{\displaystyle \gamma ^{5}}
.
Antisymétrie du produit de deux matrices gamma [ modifier | modifier le code ]
On constate aisément que quand
μ
{\displaystyle \mu }
et
ν
{\displaystyle \nu }
sont distincts on a
γ
μ
γ
ν
=
−
γ
ν
γ
μ
{\displaystyle \gamma ^{\mu }\gamma ^{\nu }=-\gamma ^{\nu }\gamma ^{\mu }}
Exemple (en représentation de Dirac)
Montrons que
γ
1
γ
2
=
−
γ
2
γ
1
{\displaystyle \gamma ^{1}\gamma ^{2}=-\gamma ^{2}\gamma ^{1}}
Ainsi
γ
1
γ
2
=
(
0
0
0
1
0
0
1
0
0
−
1
0
0
−
1
0
0
0
)
(
0
0
0
−
i
0
0
i
0
0
i
0
0
−
i
0
0
0
)
=
(
−
i
0
0
0
0
i
0
0
0
0
−
i
0
0
0
0
i
)
{\displaystyle \gamma ^{1}\gamma ^{2}={\begin{pmatrix}0&0&0&1\\0&0&1&0\\0&-1&0&0\\-1&0&0&0\end{pmatrix}}{\begin{pmatrix}0&0&0&-\mathrm {i} \\0&0&\mathrm {i} &0\\0&\mathrm {i} &0&0\\-\mathrm {i} &0&0&0\end{pmatrix}}={\begin{pmatrix}-\mathrm {i} &0&0&0\\0&\mathrm {i} &0&0\\0&0&-\mathrm {i} &0\\0&0&0&\mathrm {i} \end{pmatrix}}}
Alors que
γ
2
γ
1
=
(
0
0
0
−
i
0
0
i
0
0
i
0
0
−
i
0
0
0
)
(
0
0
0
1
0
0
1
0
0
−
1
0
0
−
1
0
0
0
)
=
(
i
0
0
0
0
−
i
0
0
0
0
i
0
0
0
0
−
i
)
{\displaystyle \gamma ^{2}\gamma ^{1}={\begin{pmatrix}0&0&0&-\mathrm {i} \\0&0&\mathrm {i} &0\\0&\mathrm {i} &0&0\\-\mathrm {i} &0&0&0\end{pmatrix}}{\begin{pmatrix}0&0&0&1\\0&0&1&0\\0&-1&0&0\\-1&0&0&0\end{pmatrix}}={\begin{pmatrix}\mathrm {i} &0&0&0\\0&-\mathrm {i} &0&0\\0&0&\mathrm {i} &0\\0&0&0&-\mathrm {i} \end{pmatrix}}}
On va maintenant généraliser l'exemple ci-dessus et le couler dans la relation d'anticommutation standard, relation vraiment fondamentale car c'est sur elle que l'algèbre de Clifford a été développée.
{
γ
μ
,
γ
ν
}
=
γ
μ
γ
ν
+
γ
ν
γ
μ
=
2
g
μ
ν
I
4
{\displaystyle \displaystyle \{\gamma ^{\mu },\gamma ^{\nu }\}=\gamma ^{\mu }\gamma ^{\nu }+\gamma ^{\nu }\gamma ^{\mu }=2g^{\mu \nu }I_{4}}
où
{
,
}
{\displaystyle \{,\}}
est le symbole de l'anticommutateur
g
μ
ν
=
g
μ
ν
{\displaystyle g^{\mu \nu }\,=g_{\mu \nu }\,}
est le tenseur fondamental ici défini au moyen de la métrique de Minkowski par
g
00
=
1
{\displaystyle g^{00}=1\,}
g
11
=
g
22
=
g
33
=
−
1
{\displaystyle g^{11}=g^{22}=g^{33}=-1\,}
g
μ
ν
=
0
{\displaystyle g^{\mu \nu }=0\,}
pour
μ
{\displaystyle \mu \,}
≠
ν
{\displaystyle \nu \,}
et
I
4
{\displaystyle \ I_{4}\,}
est la matrice identité 4x4. (le plus souvent omise)
Remarques
Par la suite quand nous parlerons de la relation standard d'anticommutation nous l'appellerons simplement relation d'anticommutation .
Comme
ν
{\displaystyle \nu }
est un indice muet on peut le remplacer dans l'expression ci-dessus par
σ
{\displaystyle \sigma }
.
Il vient :
γ
μ
γ
σ
+
γ
σ
γ
μ
=
2
g
μ
σ
I
4
{\displaystyle \displaystyle \ \gamma ^{\mu }\gamma ^{\sigma }+\gamma ^{\sigma }\gamma ^{\mu }=2g^{\mu \sigma }I_{4}}
Si on multiplie les 2 membres de cette égalité par le tenseur
g
ν
σ
{\displaystyle g_{\nu \sigma }}
on obtient
g
ν
σ
(
γ
μ
γ
σ
+
γ
σ
γ
μ
)
=
2
g
ν
σ
g
μ
σ
I
4
{\displaystyle \displaystyle \ g_{\nu \sigma }(\gamma ^{\mu }\gamma ^{\sigma }+\gamma ^{\sigma }\gamma ^{\mu })=2g_{\nu \sigma }g^{\mu \sigma }I_{4}}
Comme
g
ν
σ
{\displaystyle g_{\nu \sigma }}
est un nombre on peut le permuter avec une matrice de sorte que l'expression peut être développée comme suit:
g
ν
σ
(
γ
μ
γ
σ
+
γ
σ
γ
μ
)
=
γ
μ
g
ν
σ
γ
σ
+
g
ν
σ
γ
σ
γ
μ
=
2
g
ν
σ
g
μ
σ
I
4
{\displaystyle \displaystyle \ g_{\nu \sigma }(\gamma ^{\mu }\gamma ^{\sigma }+\gamma ^{\sigma }\gamma ^{\mu })=\gamma ^{\mu }g_{\nu \sigma }\gamma ^{\sigma }+g_{\nu \sigma }\gamma ^{\sigma }\gamma ^{\mu }=2g_{\nu \sigma }g^{\mu \sigma }I_{4}}
Si on observe que
g
μ
σ
=
g
σ
μ
{\displaystyle g^{\mu \sigma }=g^{\sigma \mu }}
et qu'on modifie dans ce sens le membre de droite
g
ν
σ
(
γ
μ
γ
σ
+
γ
σ
γ
μ
)
=
γ
μ
g
ν
σ
γ
σ
+
g
ν
σ
γ
σ
γ
μ
=
2
g
ν
σ
g
σ
μ
I
4
{\displaystyle \displaystyle \ g_{\nu \sigma }(\gamma ^{\mu }\gamma ^{\sigma }+\gamma ^{\sigma }\gamma ^{\mu })=\gamma ^{\mu }g_{\nu \sigma }\gamma ^{\sigma }+g_{\nu \sigma }\gamma ^{\sigma }\gamma ^{\mu }=2g_{\nu \sigma }g^{\sigma \mu }I_{4}}
On est alors prêts pour appliquer la règle de contraction des indices de sorte d'obtenir
γ
μ
γ
ν
+
γ
ν
γ
μ
=
2
g
ν
μ
I
4
{\displaystyle \displaystyle \ \gamma ^{\mu }\gamma _{\nu }+\gamma _{\nu }\gamma ^{\mu }=2g_{\nu }^{\mu }I_{4}}
A propos du tenseur métrique
Si l'on veut présenter le tenseur fondamental
g
μ
ν
{\displaystyle g^{\mu \nu }}
avec
μ
=
0
,
1
,
2
,
3
{\displaystyle \quad \mu =0,1,2,3\quad }
et
ν
=
0
,
1
,
2
,
3
{\displaystyle \quad \,\nu =0,1,2,3\quad }
sous sa forme matricielle générale on aura :
g
μ
ν
=
(
g
00
g
01
g
02
g
03
g
10
g
11
g
12
g
13
g
20
g
21
g
22
g
23
g
30
g
31
g
32
g
33
)
{\displaystyle g^{\mu \nu }={\begin{pmatrix}g^{00}&g^{01}&g^{02}&g^{03}\\g^{10}&g^{11}&g^{12}&g^{13}\\g^{20}&g^{21}&g^{22}&g^{23}\\g^{30}&g^{31}&g^{32}&g^{33}\end{pmatrix}}}
En le particularisant à la métrique
(
+
−
−
−
)
{\displaystyle (+---)}
on obtient
g
μ
ν
=
g
μ
ν
=
(
1
0
0
0
0
−
1
0
0
0
0
−
1
0
0
0
0
−
1
)
{\displaystyle g^{\mu \nu }=g_{\mu \nu }={\begin{pmatrix}1&0&0&0\\0&-1&0&0\\0&0&-1&0\\0&0&0&-1\end{pmatrix}}}
Ce tenseur est symétrique par suite de la symétrie du produit scalaire des vecteurs de base, on a :
g
i
j
=
g
j
i
.
{\displaystyle g_{ij}=g_{ji}\,.}
Rappelons car on utilise également ce tenseur sous cette forme dans cet article que
g
j
i
=
g
i
k
g
k
j
{\displaystyle g_{j}^{i}=g^{ik}g_{kj}\,}
où
g
j
i
{\displaystyle g_{j}^{i}\,}
sont appelées les composantes mixtes du tenseur fondamental.
Bon à savoir :
g
μ
ν
g
ν
μ
=
4
.
{\displaystyle g^{\mu \nu }g_{\nu \mu }=4\,.}
En effet cette multiplication contractée du tenseur fondamental avec lui-même contient implicitement la convention de sommation d'Einstein ce qui donne le développement suivant:
g
μ
ν
g
μ
ν
=
g
00
g
00
+
g
01
g
01
+
g
02
g
02
+
g
03
g
03
+
g
10
g
10
+
g
11
g
11
+
g
12
g
12
+
g
13
g
13
+
g
20
g
20
+
g
21
g
21
+
g
22
g
22
+
g
23
g
23
+
g
30
g
30
+
g
31
g
31
+
g
32
g
32
+
g
33
g
33
=
(
+
1
⋅
+
1
)
+
0
+
0
+
0
+
0
+
(
−
1
⋅
−
1
)
+
0
+
0
+
0
+
0
+
(
−
1
⋅
−
1
)
+
0
+
0
+
0
+
0
+
(
−
1
⋅
−
1
)
=
4
{\displaystyle {\begin{aligned}g^{\mu \nu }g_{\mu \nu }&=g^{00}g_{00}+g^{01}g_{01}+g^{02}g_{02}+g^{03}g_{03}\\&+g^{10}g_{10}+g^{11}g_{11}+g^{12}g_{12}+g^{13}g_{13}\\&+g^{20}g_{20}+g^{21}g_{21}+g^{22}g_{22}+g^{23}g_{23}\\&+g^{30}g_{30}+g^{31}g_{31}+g^{32}g_{32}+g^{33}g_{33}\\&=(+1\cdot +1)+0+0+0\\&+0+(-1\cdot -1)+0+0\\&+0+0+(-1\cdot -1)+0\\&+0+0+0+(-1\cdot -1)\\&=4\end{aligned}}}
C.Q.F.D.
Cette égalité illustre bien la notion de contraction complète sur les indices covariants et contravariants identiques dans la mesure où cette multiplication débouche ici sur le scalaire 4.
On peut remarquer qu'en utilisant la métrique
(
−
+
+
+
)
{\displaystyle (-+++)}
, soit
g
00
=
−
1
{\displaystyle g^{00}=-1\,}
g
11
=
g
22
=
g
33
=
1
.
{\displaystyle g^{11}=g^{22}=g^{33}=1\,.}
La relation d'anticommutation deviendrait :
{
γ
μ
,
γ
ν
}
=
γ
μ
γ
ν
+
γ
ν
γ
μ
=
−
2
g
μ
ν
I
4
{\displaystyle \displaystyle \{\gamma ^{\mu },\gamma ^{\nu }\}=\gamma ^{\mu }\gamma ^{\nu }+\gamma ^{\nu }\gamma ^{\mu }=-2g^{\mu \nu }I_{4}}
soit sous forme matricielle :
g
μ
ν
=
g
μ
ν
=
(
−
1
0
0
0
0
1
0
0
0
0
1
0
0
0
0
1
)
{\displaystyle g^{\mu \nu }=g_{\mu \nu }={\begin{pmatrix}-1&0&0&0\\0&1&0&0\\0&0&1&0\\0&0&0&1\end{pmatrix}}}
Il faudra donc à chaque fois se poser la question de savoir quelle métrique l'auteur d'un ouvrage/article a choisie. Dans tout cet article nous avons retenu la métrique
(
+
−
−
−
)
.
{\displaystyle (+---).}
{
γ
μ
,
γ
5
}
=
γ
μ
γ
5
+
γ
5
γ
μ
=
0
{\displaystyle \displaystyle \{\gamma ^{\mu },\gamma ^{5}\}=\gamma ^{\mu }\gamma ^{5}+\gamma ^{5}\gamma ^{\mu }=0}
ou encore
γ
μ
γ
5
=
−
γ
5
γ
μ
{\displaystyle \gamma ^{\mu }\gamma ^{5}=-\gamma ^{5}\gamma ^{\mu }}
Sur cette égalité, dont en guise de démonstration et pour reprendre les termes de W. Pauli nous ne montrerons ci-dessous qu'une spécialisation numérique , on peut voir que la propriété d'antisymétrie des produits de matrices
γ
μ
{\displaystyle \gamma ^{\mu }}
est aussi de mise avec une matrice
γ
5
{\displaystyle \gamma ^{5}}
.
Spécialisation numérique
Montrons dans la représentation de Dirac que
γ
0
γ
5
+
γ
5
γ
0
=
0.
{\displaystyle \gamma ^{0}\gamma ^{5}+\gamma ^{5}\gamma ^{0}=0.}
Ainsi
γ
0
γ
5
=
{\displaystyle \gamma ^{0}\gamma ^{5}=}
(
1
0
0
0
0
1
0
0
0
0
−
1
0
0
0
0
−
1
)
(
0
0
1
0
0
0
0
1
1
0
0
0
0
1
0
0
)
=
{\displaystyle {\begin{pmatrix}1&0&0&0\\0&1&0&0\\0&0&-1&0\\0&0&0&-1\end{pmatrix}}{\begin{pmatrix}0&0&1&0\\0&0&0&1\\1&0&0&0\\0&1&0&0\end{pmatrix}}=}
(
0
0
1
0
0
0
0
1
−
1
0
0
0
0
−
1
0
0
)
{\displaystyle {\begin{pmatrix}0&0&1&0\\0&0&0&1\\-1&0&0&0\\0&-1&0&0\end{pmatrix}}}
et
γ
5
γ
0
=
{\displaystyle \gamma ^{5}\gamma ^{0}=}
(
0
0
1
0
0
0
0
1
1
0
0
0
0
1
0
0
)
{\displaystyle {\begin{pmatrix}0&0&1&0\\0&0&0&1\\1&0&0&0\\0&1&0&0\end{pmatrix}}}
(
1
0
0
0
0
1
0
0
0
0
−
1
0
0
0
0
−
1
)
=
{\displaystyle {\begin{pmatrix}1&0&0&0\\0&1&0&0\\0&0&-1&0\\0&0&0&-1\end{pmatrix}}=}
(
0
0
−
1
0
0
0
0
−
1
1
0
0
0
0
1
0
0
)
{\displaystyle {\begin{pmatrix}0&0&-1&0\\0&0&0&-1\\1&0&0&0\\0&1&0&0\end{pmatrix}}}
La somme donne bien la matrice nulle.
En notant le commutateur par le symbole
[
,
]
{\displaystyle [,]}
, on définit couramment deux matrices d'usage pratique :
σ
μ
ν
=
i
2
[
γ
μ
,
γ
ν
]
=
i
2
(
γ
μ
γ
ν
−
γ
ν
γ
μ
)
{\displaystyle \sigma ^{\mu \nu }={\frac {\mathrm {i} }{2}}[\gamma ^{\mu },\gamma ^{\nu }]={\frac {\mathrm {i} }{2}}(\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{\nu }\gamma ^{\mu })}
S
μ
ν
=
i
4
[
γ
μ
,
γ
ν
]
=
i
4
(
γ
μ
γ
ν
−
γ
ν
γ
μ
)
{\displaystyle S^{\mu \nu }={\frac {\mathrm {i} }{4}}[\gamma ^{\mu },\gamma ^{\nu }]={\frac {\mathrm {i} }{4}}(\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{\nu }\gamma ^{\mu })}
où les matrices
S
μ
ν
{\displaystyle S^{\mu \nu }}
sont appelées les matrices de spin.
À noter les quatre relations de commutation :
[
S
ν
μ
,
S
λ
ρ
]
=
i
(
g
λ
μ
S
ν
ρ
−
g
λ
ν
S
μ
ρ
+
g
ρ
μ
S
λ
ν
−
g
ρ
ν
S
λ
μ
)
{\displaystyle \left[S^{\nu \mu },S^{\lambda \rho }\right]=\mathrm {i} (g^{\lambda \mu }S^{\nu \rho }-g^{\lambda \nu }S^{\mu \rho }+g^{\rho \mu }S^{\lambda \nu }-g^{\rho \nu }S^{\lambda \mu })}
Démonstration
On a :
[
S
ν
μ
,
S
λ
ρ
]
=
−
1
16
[
[
γ
ν
,
γ
μ
]
,
[
γ
λ
,
γ
ρ
]
]
=
−
1
16
[
2
(
g
ν
μ
I
4
−
γ
μ
γ
ν
)
,
2
(
g
λ
ρ
I
4
−
γ
ρ
γ
λ
)
]
=
−
1
4
[
g
ν
μ
I
4
−
γ
μ
γ
ν
,
g
λ
ρ
I
4
−
γ
ρ
γ
λ
]
=
−
1
4
(
[
g
ν
μ
I
4
,
g
λ
ρ
I
4
]
+
[
γ
μ
γ
ν
,
γ
ρ
γ
λ
]
−
[
γ
μ
γ
ν
,
g
λ
ρ
I
4
]
+
[
γ
ρ
γ
λ
,
g
ν
μ
I
4
]
)
{\displaystyle {\begin{aligned}\left[S^{\nu \mu },S^{\lambda \rho }\right]&=-{\frac {1}{16}}\left[\left[\gamma ^{\nu },\gamma ^{\mu }\right],\left[\gamma ^{\lambda },\gamma ^{\rho }\right]\right]\\&=-{\frac {1}{16}}\left[2\left(g^{\nu \mu }I_{4}-\gamma ^{\mu }\gamma ^{\nu }\right),2\left(g^{\lambda \rho }I_{4}-\gamma ^{\rho }\gamma ^{\lambda }\right)\right]\\&=-{\frac {1}{4}}\left[g^{\nu \mu }I_{4}-\gamma ^{\mu }\gamma ^{\nu },g^{\lambda \rho }I_{4}-\gamma ^{\rho }\gamma ^{\lambda }\right]\\&=-{\frac {1}{4}}\left(\left[g^{\nu \mu }I_{4},g^{\lambda \rho }I_{4}\right]+\left[\gamma ^{\mu }\gamma ^{\nu },\gamma ^{\rho }\gamma ^{\lambda }\right]-\left[\gamma ^{\mu }\gamma ^{\nu },g^{\lambda \rho }I_{4}\right]+\left[\gamma ^{\rho }\gamma ^{\lambda },g^{\nu \mu }I_{4}\right]\right)\end{aligned}}}
À ce stade, on observe dans le membre de droite, que dans les premier, troisième et quatrième commutateurs, la matrice 4x4 constituant l'un des termes au moins ne prend que des valeurs scalaires 1 ou -1, ce qui implique la nullité de ces commutateurs. Il vient donc :
[
S
ν
μ
,
S
λ
ρ
]
=
−
1
4
[
γ
μ
γ
ν
,
γ
ρ
γ
λ
]
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
γ
ρ
γ
λ
γ
μ
γ
ν
)
{\displaystyle {\begin{aligned}\left[S^{\nu \mu },S^{\lambda \rho }\right]&=-{\frac {1}{4}}\left[\gamma ^{\mu }\gamma ^{\nu },\gamma ^{\rho }\gamma ^{\lambda }\right]\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-\gamma ^{\rho }\gamma ^{\lambda }\gamma ^{\mu }\gamma ^{\nu }\right)\end{aligned}}}
On va maintenant se faire téléscoper les différents termes en utilisant la relation anti-commutation :
{
γ
μ
,
γ
ν
}
=
γ
μ
γ
ν
+
γ
ν
γ
μ
=
2
g
μ
ν
I
4
{\displaystyle \{\gamma ^{\mu },\gamma ^{\nu }\}=\gamma ^{\mu }\gamma ^{\nu }+\gamma ^{\nu }\gamma ^{\mu }=2g^{\mu \nu }I_{4}}
ce qui implique :
γ
μ
γ
ν
=
2
g
μ
ν
I
4
−
γ
ν
γ
μ
{\displaystyle \gamma ^{\mu }\gamma ^{\nu }=2g^{\mu \nu }I_{4}-\gamma ^{\nu }\gamma ^{\mu }}
jusqu'à faire apparaître un terme qui se compensera avec le premier terme du membre de droite.
[
S
ν
μ
,
S
λ
ρ
]
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
γ
ρ
(
2
g
λ
μ
I
4
−
γ
μ
γ
λ
)
γ
ν
)
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
2
g
λ
μ
I
4
γ
ρ
γ
ν
+
γ
ρ
γ
μ
γ
λ
γ
ν
)
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
2
g
λ
μ
I
4
γ
ρ
γ
ν
+
γ
ρ
γ
μ
(
2
g
λ
ν
I
4
−
γ
ν
γ
λ
)
)
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
2
g
λ
μ
I
4
γ
ρ
γ
ν
+
2
g
λ
ν
I
4
γ
ρ
γ
μ
−
γ
ρ
γ
μ
γ
ν
γ
λ
)
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
2
g
λ
μ
I
4
γ
ρ
γ
ν
+
2
g
λ
ν
I
4
γ
ρ
γ
μ
−
(
2
g
ρ
μ
I
4
−
γ
μ
γ
ρ
)
γ
ν
γ
λ
)
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
2
g
λ
μ
I
4
γ
ρ
γ
ν
+
2
g
λ
ν
I
4
γ
ρ
γ
μ
−
2
g
ρ
μ
I
4
γ
ν
γ
λ
+
γ
μ
γ
ρ
γ
ν
γ
λ
)
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
2
g
λ
μ
I
4
γ
ρ
γ
ν
+
2
g
λ
ν
I
4
γ
ρ
γ
μ
−
2
g
ρ
μ
I
4
γ
ν
γ
λ
+
γ
μ
(
2
g
ρ
ν
I
4
−
γ
ν
γ
ρ
)
γ
λ
)
=
−
1
4
(
γ
μ
γ
ν
γ
ρ
γ
λ
−
2
g
λ
μ
I
4
γ
ρ
γ
ν
+
2
g
λ
ν
I
4
γ
ρ
γ
μ
−
2
g
ρ
μ
I
4
γ
ν
γ
λ
+
2
g
ρ
ν
I
4
γ
μ
γ
λ
−
γ
μ
γ
ν
γ
ρ
γ
λ
)
=
1
2
(
g
λ
μ
I
4
γ
ρ
γ
ν
−
g
λ
ν
I
4
γ
ρ
γ
μ
+
g
ρ
μ
I
4
γ
ν
γ
λ
−
g
ρ
ν
I
4
γ
μ
γ
λ
)
=
1
i
g
λ
μ
I
4
(
i
γ
ρ
γ
ν
/
2
)
−
1
i
g
λ
ν
I
4
(
i
γ
ρ
γ
μ
/
2
)
+
1
i
g
ρ
μ
I
4
(
i
γ
ν
γ
λ
/
2
)
−
1
i
g
ρ
ν
I
4
(
i
γ
μ
γ
λ
/
2
)
{\displaystyle {\begin{aligned}\left[S^{\nu \mu },S^{\lambda \rho }\right]&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-\gamma ^{\rho }\left(2g^{\lambda \mu }I_{4}-\gamma ^{\mu }\gamma ^{\lambda }\right)\gamma ^{\nu }\right)\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-2g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }+\gamma ^{\rho }\gamma ^{\mu }\gamma ^{\lambda }\gamma ^{\nu }\right)\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-2g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }+\gamma ^{\rho }\gamma ^{\mu }\left(2g^{\lambda \nu }I_{4}-\gamma ^{\nu }\gamma ^{\lambda }\right)\right)\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-2g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }+2g^{\lambda \nu }I_{4}\gamma ^{\rho }\gamma ^{\mu }-\gamma ^{\rho }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\lambda }\right)\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-2g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }+2g^{\lambda \nu }I_{4}\gamma ^{\rho }\gamma ^{\mu }-\left(2g^{\rho \mu }I_{4}-\gamma ^{\mu }\gamma ^{\rho }\right)\gamma ^{\nu }\gamma ^{\lambda }\right)\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-2g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }+2g^{\lambda \nu }I_{4}\gamma ^{\rho }\gamma ^{\mu }-2g^{\rho \mu }I_{4}\gamma ^{\nu }\gamma ^{\lambda }+\gamma ^{\mu }\gamma ^{\rho }\gamma ^{\nu }\gamma ^{\lambda }\right)\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-2g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }+2g^{\lambda \nu }I_{4}\gamma ^{\rho }\gamma ^{\mu }-2g^{\rho \mu }I_{4}\gamma ^{\nu }\gamma ^{\lambda }+\gamma ^{\mu }\left(2g^{\rho \nu }I_{4}-\gamma ^{\nu }\gamma ^{\rho }\right)\gamma ^{\lambda }\right)\\&=-{\frac {1}{4}}\left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }-2g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }+2g^{\lambda \nu }I_{4}\gamma ^{\rho }\gamma ^{\mu }-2g^{\rho \mu }I_{4}\gamma ^{\nu }\gamma ^{\lambda }+2g^{\rho \nu }I_{4}\gamma ^{\mu }\gamma ^{\lambda }-\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\lambda }\right)\\&={\frac {1}{2}}\left(g^{\lambda \mu }I_{4}\gamma ^{\rho }\gamma ^{\nu }-g^{\lambda \nu }I_{4}\gamma ^{\rho }\gamma ^{\mu }+g^{\rho \mu }I_{4}\gamma ^{\nu }\gamma ^{\lambda }-g^{\rho \nu }I_{4}\gamma ^{\mu }\gamma ^{\lambda }\right)\\&={\frac {1}{\mathrm {i} }}g^{\lambda \mu }I_{4}\left(i\gamma ^{\rho }\gamma ^{\nu }/2\right)-{\frac {1}{\mathrm {i} }}g^{\lambda \nu }I_{4}\left(\mathrm {i} \gamma ^{\rho }\gamma ^{\mu }/2\right)+{\frac {1}{\mathrm {i} }}g^{\rho \mu }I_{4}\left(\mathrm {i} \gamma ^{\nu }\gamma ^{\lambda }/2\right)-{\frac {1}{\mathrm {i} }}g^{\rho \nu }I_{4}\left(i\gamma ^{\mu }\gamma ^{\lambda }/2\right)\end{aligned}}}
Pour achever la démonstration nous aurons besoin de la relation :
i
2
γ
ρ
γ
ν
=
i
2
g
ν
ρ
−
S
ν
ρ
{\displaystyle {\frac {\mathrm {i} }{2}}\gamma ^{\rho }\gamma {\nu }={\frac {\mathrm {i} }{2}}g^{\nu \rho }-S^{\nu \rho }}
Celle-ci se démontre aisément en partant d'une part de la relation d'anticommutation :
2
g
μ
ν
=
γ
μ
γ
ν
+
γ
ν
γ
μ
{\displaystyle 2g^{\mu \nu }=\gamma ^{\mu }\gamma ^{\nu }+\gamma ^{\nu }\gamma ^{\mu }}
et d'autre part de la définition
S
μ
ν
=
i
4
γ
μ
γ
ν
−
i
4
γ
ν
γ
μ
{\displaystyle S^{\mu \nu }={\frac {\mathrm {i} }{4}}\gamma ^{\mu }\gamma ^{\nu }-{\frac {\mathrm {i} }{4}}\gamma ^{\nu }\gamma ^{\mu }}
car il suffit de multiplier la première par i/4 et d'en soustraire la deuxième.
On peut ainsi conclure le calcul :
[
S
ν
μ
,
S
λ
ρ
]
=
1
i
g
λ
μ
I
4
(
i
γ
ρ
γ
ν
/
2
)
−
1
i
g
λ
ν
I
4
(
i
γ
ρ
γ
μ
/
2
)
+
1
i
g
ρ
μ
I
4
(
i
γ
ν
γ
λ
/
2
)
−
1
i
g
ρ
ν
I
4
(
i
γ
μ
γ
λ
/
2
)
=
1
i
g
λ
μ
I
4
(
i
2
g
ν
ρ
−
S
ν
ρ
)
−
1
i
g
λ
ν
I
4
(
i
2
g
μ
ρ
−
S
μ
ρ
)
+
1
i
g
ρ
μ
I
4
(
i
2
g
λ
ν
−
S
λ
ν
)
−
1
i
g
ρ
ν
I
4
(
i
2
g
λ
μ
−
S
λ
μ
)
=
i
(
g
λ
μ
S
ν
ρ
−
g
λ
ν
S
μ
ρ
+
g
ρ
μ
S
λ
ν
−
g
ρ
ν
S
λ
μ
)
{\displaystyle {\begin{aligned}\left[S^{\nu \mu },S^{\lambda \rho }\right]&={\frac {1}{\mathrm {i} }}g^{\lambda \mu }I_{4}\left(\mathrm {i} \gamma ^{\rho }\gamma ^{\nu }/2\right)-{\frac {1}{\mathrm {i} }}g^{\lambda \nu }I_{4}\left(i\gamma ^{\rho }\gamma ^{\mu }/2\right)+{\frac {1}{\mathrm {i} }}g^{\rho \mu }I_{4}\left(i\gamma ^{\nu }\gamma ^{\lambda }/2\right)-{\frac {1}{\mathrm {i} }}g^{\rho \nu }I_{4}\left(\mathrm {i} \gamma ^{\mu }\gamma ^{\lambda }/2\right)\\&={\frac {1}{\mathrm {i} }}g^{\lambda \mu }I_{4}\left({\frac {\mathrm {i} }{2}}g^{\nu \rho }-S^{\nu \rho }\right)-{\frac {1}{\mathrm {i} }}g^{\lambda \nu }I_{4}\left({\frac {\mathrm {i} }{2}}g^{\mu \rho }-S^{\mu \rho }\right)+{\frac {1}{\mathrm {i} }}g^{\rho \mu }I_{4}\left({\frac {\mathrm {i} }{2}}g^{\lambda \nu }-S^{\lambda \nu }\right)-{\frac {1}{\mathrm {i} }}g^{\rho \nu }I_{4}\left({\frac {i}{2}}g^{\lambda \mu }-S^{\lambda \mu }\right)\\&=\mathrm {i} \left(g^{\lambda \mu }S^{\nu \rho }-g^{\lambda \nu }S^{\mu \rho }+g^{\rho \mu }S^{\lambda \nu }-g^{\rho \nu }S^{\lambda \mu }\right)\end{aligned}}}
C.Q.F.D.
[
σ
μ
ν
,
σ
α
β
]
=
2
i
(
g
μ
β
σ
ν
α
+
g
ν
α
σ
μ
β
−
g
μ
α
σ
ν
β
−
g
ν
β
σ
μ
α
)
{\displaystyle \left[\sigma ^{\mu \nu },\sigma ^{\alpha \beta }\right]=2\mathrm {i} (g^{\mu \beta }\sigma ^{\nu \alpha }+g^{\nu \alpha }\sigma ^{\mu \beta }-g^{\mu \alpha }\sigma ^{\nu \beta }-g^{\nu \beta }\sigma ^{\mu \alpha })}
Démonstration
La démonstration est identique à celle donnée ci-dessus.
C.Q.F.D.
[
γ
5
,
σ
μ
ν
]
=
0
{\displaystyle \left[\gamma ^{5},\sigma ^{\mu \nu }\right]=0}
Démonstration
En effet
[
γ
5
,
σ
μ
ν
]
=
γ
5
i
2
(
γ
μ
γ
ν
−
γ
ν
γ
μ
)
−
i
2
(
γ
μ
γ
ν
−
γ
ν
γ
μ
)
γ
5
=
i
2
(
γ
5
γ
μ
γ
ν
−
γ
5
γ
ν
γ
μ
−
γ
μ
γ
ν
γ
5
+
γ
ν
γ
μ
γ
5
)
=
i
2
(
γ
5
γ
μ
γ
ν
−
γ
5
γ
ν
γ
μ
−
γ
5
γ
μ
γ
ν
+
γ
5
γ
ν
γ
μ
)
=
0
{\displaystyle {\begin{aligned}\left[\gamma ^{5},\sigma ^{\mu \nu }\right]&=\gamma ^{5}{\frac {\mathrm {i} }{2}}(\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{\nu }\gamma ^{\mu })-{\frac {\mathrm {i} }{2}}(\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{\nu }\gamma ^{\mu })\gamma ^{5}\\&={\frac {\mathrm {\mathrm {i} } }{2}}(\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{5}\gamma ^{\nu }\gamma ^{\mu }-\gamma ^{\mu }\gamma ^{\nu }\gamma ^{5}+\gamma ^{\nu }\gamma ^{\mu }\gamma ^{5})\\&={\frac {\mathrm {\mathrm {i} } }{2}}(\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{5}\gamma ^{\nu }\gamma ^{\mu }-\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }+\gamma ^{5}\gamma ^{\nu }\gamma ^{\mu })\\&=0\end{aligned}}}
C.Q.F.D.
[
γ
5
,
S
μ
ν
]
=
0
{\displaystyle \left[\gamma ^{5},S^{\mu \nu }\right]=0}
Démonstration
Etant donnée la proportionnalité entre
S
μ
ν
{\displaystyle S^{\mu \nu }}
et
σ
μ
ν
{\displaystyle \sigma ^{\mu \nu }}
la démonstration est identique à celle donnée ci-dessus.
C.Q.F.D.
Les matrices gamma font l'objet de propriétés d'hermiticité telles que les relations d'anticommutation soient respectées.
Num
Identité propre
1
γ
0
=
γ
0
{\displaystyle \displaystyle \gamma _{0}=\gamma ^{0}}
2
(
γ
0
)
2
=
I
4
{\displaystyle \displaystyle (\gamma ^{0})^{2}=I_{4}}
3
γ
i
=
−
γ
i
p
o
u
r
i
=
1
,
2
,
3
{\displaystyle \displaystyle \gamma _{i}=-\gamma ^{i}\quad \mathrm {pour} \quad i=1,2,3}
4
(
γ
i
)
2
=
−
I
4
p
o
u
r
i
=
1
,
2
,
3
{\displaystyle \displaystyle (\gamma ^{i})^{2}=-I_{4}\quad \mathrm {pour} \quad i=1,2,3}
5
γ
5
=
γ
5
=
−
i
γ
0
γ
1
γ
2
γ
3
{\displaystyle \displaystyle \gamma _{5}=\gamma ^{5}=-\mathrm {i} \gamma _{0}\gamma _{1}\gamma _{2}\gamma _{3}}
6
(
γ
5
)
2
=
I
4
{\displaystyle \displaystyle (\gamma ^{5})^{2}=I_{4}}
7
(
γ
0
)
†
=
γ
0
{\displaystyle \displaystyle (\gamma ^{0})^{\dagger }=\gamma ^{0}}
8
(
γ
i
)
†
=
−
γ
i
p
o
u
r
i
=
1
,
2
,
3
{\displaystyle \displaystyle (\gamma ^{i})^{\dagger }=-\gamma ^{i}\quad \mathrm {pour} \quad i=1,2,3}
9
(
γ
μ
)
†
=
γ
0
γ
μ
γ
0
p
o
u
r
μ
=
0
,
1
,
2
,
3
{\displaystyle \displaystyle (\gamma ^{\mu })^{\dagger }=\gamma ^{0}\gamma ^{\mu }\gamma ^{0}\quad \mathrm {pour} \quad \mu =0,1,2,3}
10
σ
μ
ν
=
−
σ
ν
μ
{\displaystyle \displaystyle \sigma ^{\mu \nu }=-\sigma ^{\nu \mu }}
Les identités qui suivent découlent des relations fondamentales d'anticommutation ainsi que des identités propres, de sorte qu'elles sont valables dans n'importe quelle base ou représentation.
Num
Identité de contraction
1
γ
μ
γ
μ
=
4
I
4
{\displaystyle \displaystyle \gamma ^{\mu }\gamma _{\mu }=4I_{4}}
2
γ
μ
γ
ν
γ
μ
=
−
2
γ
ν
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma _{\mu }=-2\gamma ^{\nu }}
3
γ
μ
γ
ν
γ
ρ
γ
μ
=
4
g
ν
ρ
I
4
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma _{\mu }=4g^{\nu \rho }I_{4}}
4
γ
μ
γ
ν
γ
ρ
γ
σ
γ
μ
=
−
2
γ
σ
γ
ρ
γ
ν
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma _{\mu }=-2\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\nu }}
Démonstration
Preuve de l'identité no 1
Pour démontrer que
γ
μ
γ
μ
=
4
I
4
{\displaystyle \gamma ^{\mu }\gamma _{\mu }=4I_{4}}
l'idée est de se servir de la relation d'anticommutation.
L'utilisation de la métrique g permet de transformer la contraction
γ
μ
γ
μ
{\displaystyle \gamma ^{\mu }\gamma _{\mu }}
de manière à pouvoir faire appel à cette relation d'anticommutation :
γ
μ
γ
μ
=
γ
μ
g
μ
ν
γ
ν
=
g
μ
ν
γ
μ
γ
ν
=
1
2
(
g
μ
ν
+
g
ν
μ
)
γ
μ
γ
ν
(
s
y
m
e
t
r
i
e
d
e
g
)
=
1
2
(
g
μ
ν
γ
μ
γ
ν
+
g
ν
μ
γ
μ
γ
ν
)
(
e
x
p
a
n
s
i
o
n
)
=
1
2
(
g
μ
ν
γ
μ
γ
ν
+
g
μ
ν
γ
ν
γ
μ
)
(
n
o
u
v
e
l
l
e
l
a
b
e
l
l
i
s
a
t
i
o
n
d
u
t
e
r
m
e
d
e
d
r
o
i
t
e
)
=
1
2
g
μ
ν
{
γ
μ
,
γ
ν
}
=
1
2
g
μ
ν
(
2
g
μ
ν
I
4
)
(
r
e
l
a
t
i
o
n
d
′
a
n
t
i
c
o
m
m
u
t
a
t
i
o
n
)
=
g
μ
ν
g
μ
ν
I
4
=
4
I
4
.
{\displaystyle {\begin{matrix}\gamma ^{\mu }\gamma _{\mu }&=\gamma ^{\mu }g_{\mu \nu }\gamma ^{\nu }=g_{\mu \nu }\gamma ^{\mu }\gamma ^{\nu }&\\&={\frac {1}{2}}(g_{\mu \nu }+g_{\nu \mu })\gamma ^{\mu }\gamma ^{\nu }&\mathrm {(symetrie\ de} \ g)\\&={\frac {1}{2}}(g_{\mu \nu }\gamma ^{\mu }\gamma ^{\nu }+g_{\nu \mu }\gamma ^{\mu }\gamma ^{\nu })&\mathrm {(expansion)} \\&={\frac {1}{2}}(g_{\mu \nu }\gamma ^{\mu }\gamma ^{\nu }+g_{\mu \nu }\gamma ^{\nu }\gamma ^{\mu })&\mathrm {(nouvelle\ labellisation\ du\ terme\ de\ droite)} \\&={\frac {1}{2}}g_{\mu \nu }\{\gamma ^{\mu },\gamma ^{\nu }\}&\\&={\frac {1}{2}}g_{\mu \nu }\left(2g^{\mu \nu }I_{4}\right)&\mathrm {(relation\ d'anticommutation)} \\&=g_{\mu \nu }g^{\mu \nu }I_{4}=4I_{4}.&\end{matrix}}}
C.Q.F.D.
Preuve de l'identité no 2
On a :
γ
μ
γ
ν
γ
μ
=
γ
μ
(
2
g
μ
ν
I
4
−
γ
μ
γ
ν
)
=
2
γ
μ
g
μ
ν
−
γ
μ
γ
μ
γ
ν
=
2
γ
ν
−
4
γ
ν
=
−
2
γ
ν
.
{\displaystyle {\begin{aligned}\gamma ^{\mu }\gamma ^{\nu }\gamma _{\mu }&=\gamma ^{\mu }\left(2g_{\mu }^{\nu }I_{4}-\gamma _{\mu }\gamma ^{\nu }\right)\\&=2\gamma ^{\mu }g_{\mu }^{\nu }-\gamma ^{\mu }\gamma _{\mu }\gamma ^{\nu }\\&=2\gamma ^{\nu }-4\gamma ^{\nu }=-2\gamma ^{\nu }\end{aligned}}.}
C.Q.F.D.
Preuve de l'identité no 3
Pour démontrer que
γ
μ
γ
ν
γ
ρ
γ
μ
=
4
g
ν
ρ
I
4
{\displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma _{\mu }=4g^{\nu \rho }I_{4}\,}
on utilise deux fois la relation d'anticommutation pour déplacer
γ
μ
{\displaystyle \gamma ^{\mu }}
vers la droite
γ
μ
γ
ν
γ
ρ
γ
μ
=
{
γ
μ
,
γ
ν
}
γ
ρ
γ
μ
−
γ
ν
γ
μ
γ
ρ
γ
μ
=
2
g
μ
ν
γ
ρ
γ
μ
−
γ
ν
{
γ
μ
,
γ
ρ
}
γ
μ
+
γ
ν
γ
ρ
γ
μ
γ
μ
.
{\displaystyle {\begin{aligned}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma _{\mu }&=\{\gamma ^{\mu },\gamma ^{\nu }\}\gamma ^{\rho }\gamma _{\mu }-\gamma ^{\nu }\gamma ^{\mu }\gamma ^{\rho }\gamma _{\mu }\\&=2\ g^{\mu \nu }\gamma ^{\rho }\gamma _{\mu }-\gamma ^{\nu }\{\gamma ^{\mu },\gamma ^{\rho }\}\gamma _{\mu }+\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\mu }\gamma _{\mu }.\end{aligned}}}
Comme
γ
μ
γ
μ
=
4
I
4
{\displaystyle \gamma ^{\mu }\gamma _{\mu }=4I_{4}\,}
on peut contracter les deux dernières matrices gamma dans le membre de droite de l'expression ci-dessus, de sorte à obtenir
γ
μ
γ
ν
γ
ρ
γ
μ
=
2
γ
ρ
γ
ν
−
γ
ν
2
g
μ
ρ
γ
μ
+
4
γ
ν
γ
ρ
=
2
γ
ρ
γ
ν
−
2
γ
ν
γ
ρ
+
4
γ
ν
γ
ρ
=
2
(
γ
ρ
γ
ν
+
γ
ν
γ
ρ
)
=
2
{
γ
ν
,
γ
ρ
}
.
{\displaystyle {\begin{aligned}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma _{\mu }&=2\gamma ^{\rho }\gamma ^{\nu }-\gamma ^{\nu }2g^{\mu \rho }\gamma _{\mu }+4\ \gamma ^{\nu }\gamma ^{\rho }\\&=2\gamma ^{\rho }\gamma ^{\nu }-2\ \gamma ^{\nu }\gamma ^{\rho }+4\ \gamma ^{\nu }\gamma ^{\rho }\\&=2(\gamma ^{\rho }\gamma ^{\nu }+\gamma ^{\nu }\gamma ^{\rho })\\&=2\{\gamma ^{\nu },\gamma ^{\rho }\}.\end{aligned}}}
Finalement, en faisant jouer une dernière fois la relation d'anticommutation, il vient
γ
μ
γ
ν
γ
ρ
γ
μ
=
4
g
ν
ρ
I
4
.
{\displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma _{\mu }=4\ g^{\nu \rho }I_{4}\,.}
C.Q.F.D.
Preuve de l'identité no 4
Pour démontrer que
γ
μ
γ
ν
γ
ρ
γ
σ
γ
μ
=
−
2
γ
σ
γ
ρ
γ
ν
{\displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma _{\mu }=-2\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\nu }}
on procède comme suit :
γ
μ
γ
ν
γ
ρ
γ
σ
γ
μ
=
(
2
g
μ
ν
−
γ
ν
γ
μ
)
γ
ρ
γ
σ
γ
μ
(
r
e
l
a
t
i
o
n
d
′
a
n
t
i
c
o
m
m
u
t
a
t
i
o
n
s
u
r
γ
μ
γ
ν
)
=
2
g
μ
ν
γ
ρ
γ
σ
γ
μ
−
4
γ
ν
g
ρ
σ
(
u
t
i
l
i
s
a
t
i
o
n
d
e
l
a
t
r
o
i
s
i
e
m
e
i
d
e
n
t
i
t
e
d
e
c
o
n
t
r
a
c
t
i
o
n
γ
μ
γ
ρ
γ
σ
γ
μ
=
4
g
ρ
σ
I
4
)
=
2
γ
ρ
γ
σ
γ
ν
−
4
γ
ν
g
ρ
σ
(
u
t
i
l
i
s
a
t
i
o
n
d
e
l
a
m
e
t
r
i
q
u
e
p
o
u
r
m
o
n
t
e
r
l
′
i
n
d
i
c
e
μ
)
=
2
(
2
g
ρ
σ
−
γ
σ
γ
ρ
)
γ
ν
−
4
γ
ν
g
ρ
σ
(
r
e
l
a
t
i
o
n
d
′
a
n
t
i
c
o
m
m
u
t
a
t
i
o
n
s
u
r
γ
ρ
γ
σ
)
=
−
2
γ
σ
γ
ρ
γ
ν
(
p
a
r
s
i
m
p
l
i
f
i
c
a
t
i
o
n
d
e
d
e
u
x
t
e
r
m
e
s
e
n
t
e
n
a
n
t
c
o
m
p
t
e
d
u
f
a
i
t
q
u
e
g
ρ
σ
γ
ν
=
γ
ν
g
ρ
σ
)
{\displaystyle {\begin{matrix}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma _{\mu }&=(2g^{\mu \nu }-\gamma ^{\nu }\gamma ^{\mu })\gamma ^{\rho }\gamma ^{\sigma }\gamma _{\mu }&\mathrm {(relation\ d'anticommutation\ sur} \gamma ^{\mu }\gamma ^{\nu })\\&=2g^{\mu \nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma _{\mu }-4\gamma ^{\nu }g^{\rho \sigma }&\mathrm {(utilisation\ de\ la\ troisieme\ identite\ de\ contraction} \gamma ^{\mu }\gamma ^{\rho }\gamma ^{\sigma }\gamma _{\mu }=4g^{\rho \sigma }I_{4})\\&=2\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\nu }-4\gamma ^{\nu }g^{\rho \sigma }&\mathrm {(utilisation\ de\ la\ metrique\ pour\ monter\ l'indice} \mu )\\&=2(2g^{\rho \sigma }-\gamma ^{\sigma }\gamma ^{\rho })\gamma ^{\nu }-4\gamma ^{\nu }g^{\rho \sigma }&\mathrm {(relation\ d'anticommutation\ sur} \gamma ^{\rho }\gamma ^{\sigma })\\&=-2\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\nu }&\mathrm {(par\ simplification\ de\ deux\ termes\ en\ tenant\ compte\ du\ fait\ que} g^{\rho \sigma }\gamma ^{\nu }=\gamma ^{\nu }g^{\rho \sigma })\end{matrix}}}
C.Q.F.D.
Identités de produits de matrices différentes [ modifier | modifier le code ]
Num
Identité de produit
1
γ
μ
γ
ν
=
g
μ
ν
−
i
σ
μ
ν
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }=g^{\mu \nu }-\mathrm {i} \sigma ^{\mu \nu }}
2
γ
μ
γ
ν
γ
ρ
=
g
μ
ν
γ
ρ
+
g
ν
ρ
γ
μ
−
g
μ
ρ
γ
ν
−
i
ϵ
σ
μ
ν
ρ
γ
σ
γ
5
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }=g^{\mu \nu }\gamma ^{\rho }+g^{\nu \rho }\gamma ^{\mu }-g^{\mu \rho }\gamma ^{\nu }-\mathrm {i} \epsilon ^{\sigma \mu \nu \rho }\gamma _{\sigma }\gamma ^{5}}
3
γ
μ
γ
ν
γ
ρ
=
2
γ
μ
g
ν
ρ
+
2
γ
ρ
g
μ
ν
−
2
γ
ν
g
μ
ρ
−
γ
ν
γ
μ
γ
ρ
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }=2\gamma ^{\mu }g^{\nu \rho }+2\gamma ^{\rho }g^{\mu \nu }-2\gamma ^{\nu }g^{\mu \rho }-\gamma ^{\nu }\gamma ^{\mu }\gamma ^{\rho }}
Démonstration
Preuve de l'identité no 1
On va développer la matrice sigma d'une part et appliquer la relation d'anticommutation d'autre part.
Rappelons que
σ
μ
ν
=
i
2
[
γ
μ
,
γ
ν
]
=
i
2
(
γ
μ
γ
ν
−
γ
ν
γ
μ
)
.
{\displaystyle \sigma ^{\mu \nu }={\frac {i}{2}}[\gamma ^{\mu },\gamma ^{\nu }]={\frac {\mathrm {i} }{2}}(\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{\nu }\gamma ^{\mu }).}
Dès lors le membre de droite de l'identité devient :
g
μ
ν
−
i
σ
μ
ν
=
g
μ
ν
+
1
2
γ
μ
γ
ν
−
1
2
γ
ν
γ
μ
.
{\displaystyle \displaystyle g^{\mu \nu }-\mathrm {i} \sigma ^{\mu \nu }=g^{\mu \nu }+{\frac {1}{2}}\gamma ^{\mu }\gamma ^{\nu }-{\frac {1}{2}}\gamma ^{\nu }\gamma ^{\mu }.}
Et en reportant ceci dans la relation à démontrer et avoir réarrangé les termes il vient:
g
μ
ν
=
1
2
γ
μ
γ
ν
+
1
2
γ
ν
γ
μ
{\displaystyle \displaystyle g^{\mu \nu }={\frac {1}{2}}\gamma ^{\mu }\gamma ^{\nu }+{\frac {1}{2}}\gamma ^{\nu }\gamma ^{\mu }}
de sorte qu'on retrouve la relation d'anticommutation.
C.Q.F.D.
Preuve de l'identité no 2
Pour démontrer que
γ
μ
γ
ν
γ
ρ
=
g
μ
ν
γ
ρ
+
g
ν
ρ
γ
μ
−
g
μ
ρ
γ
ν
−
i
ϵ
σ
μ
ν
ρ
γ
σ
γ
5
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }=g^{\mu \nu }\gamma ^{\rho }+g^{\nu \rho }\gamma ^{\mu }-g^{\mu \rho }\gamma ^{\nu }-\mathrm {i} \epsilon ^{\sigma \mu \nu \rho }\gamma _{\sigma }\gamma ^{5}}
on va faire une analyse exhaustive des cas susceptibles de se produire.
Cas 1 : les trois indices
μ
,
ν
,
ρ
{\displaystyle \mu ,\nu ,\rho }
sont égaux
L'égalité devient par exemple
γ
μ
γ
μ
γ
μ
=
g
μ
μ
γ
μ
+
g
μ
μ
γ
μ
−
g
μ
μ
γ
μ
−
i
ϵ
σ
μ
μ
μ
γ
σ
γ
5
.
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\mu }\gamma ^{\mu }=g^{\mu \mu }\gamma ^{\mu }+g^{\mu \mu }\gamma ^{\mu }-g^{\mu \mu }\gamma ^{\mu }-\mathrm {i} \epsilon ^{\sigma \mu \mu \mu }\gamma _{\sigma }\gamma ^{5}.}
Dans ce cas, moyennant les identités propres 2 et 4, le membre de gauche vaut +/-
γ
μ
.
{\displaystyle \gamma ^{\mu }\,.}
Dans le membre de droite on note que deux termes se compensent et que par ailleurs le symbole de Levi-Civita vaut 0 de sorte qu'on se retrouve avec un seul terme à savoir
+
g
μ
μ
γ
μ
.
{\displaystyle +g^{\mu \mu }\gamma ^{\mu }.}
Si
μ
=
0
{\displaystyle \mu =0}
alors
γ
μ
=
γ
μ
{\displaystyle \,\gamma ^{\mu }=\gamma ^{\mu }}
Si
μ
=
1
,
2
ou
3
{\displaystyle \mu =1,2\ {\textrm {ou}}\ 3\,}
alors
−
γ
μ
=
−
γ
μ
{\displaystyle \,-\gamma ^{\mu }=-\gamma ^{\mu }}
Cas 2 : deux indices sont égaux et différents du troisième
L'égalité devient par exemple
γ
μ
γ
ν
γ
ν
=
g
μ
ν
γ
ν
+
g
ν
ν
γ
μ
−
g
μ
ν
γ
ν
−
i
ϵ
σ
μ
ν
ν
γ
σ
γ
5
.
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\nu }=g^{\mu \nu }\gamma ^{\nu }+g^{\nu \nu }\gamma ^{\mu }-g^{\mu \nu }\gamma ^{\nu }-\mathrm {i} \epsilon ^{\sigma \mu \nu \nu }\gamma _{\sigma }\gamma ^{5}.}
Dans ce cas le membre de gauche vaut
±
γ
μ
.
{\displaystyle \pm \gamma ^{\mu }\,.}
Dans le membre de droite on note que le symbole de Levi-Civita vaut 0 et par ailleurs
g
μ
ν
=
0.
{\displaystyle g^{\mu \nu }=0.}
Or
g
ν
ν
=
+
/
−
1.
{\displaystyle g^{\nu \nu }=+/-1.}
Dès lors l'égalité est évidente.
Cas 3 : les trois indices
μ
,
ν
,
ρ
{\displaystyle \mu ,\nu ,\rho }
sont différents les uns des autres
Examinons le membre de droite :
les trois termes comportant un facteur g sont nuls;
ne reste donc que le terme
−
i
ϵ
σ
μ
ν
ρ
γ
σ
γ
5
.
{\displaystyle \displaystyle -\mathrm {i} \epsilon ^{\sigma \mu \nu \rho }\gamma _{\sigma }\gamma ^{5}\,.}
La sommation sur
σ
{\displaystyle \sigma }
comporte quatre termes ; quand
σ
{\displaystyle \sigma }
est égal à l'une des valeurs prises par
μ
,
ν
{\displaystyle \mu ,\nu }
ou
ρ
{\displaystyle \rho }
le symbole de Levi-Civita sera nul. Des quatre termes de la somme, il n'en reste dès lors plus qu'un seul celui pour lequel les quatre indices sont différents.
Dans ce cas le symbole prendra la valeur +/- 1. En se souvenant que, par les identités propres 1 et 3,
γ
σ
=
+
/
−
γ
σ
{\displaystyle \gamma _{\sigma }=+/-\gamma ^{\sigma }}
on voit qu'on se trouve dans les conditions d'application du lemme 1 de Pauli (avec
γ
A
=
γ
σ
{\displaystyle \gamma ^{A}=\gamma ^{\sigma }}
,
γ
B
=
γ
5
{\displaystyle \gamma ^{B}=\gamma ^{5}}
et
γ
C
=
i
γ
μ
γ
ν
γ
ρ
{\displaystyle \gamma ^{C}=\mathrm {i} \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }}
)
C.Q.F.D.
Les trois premières identités ci-dessous sont en fait des définitions indépendantes les unes des autres, aucune d'entre elles ne se déduisant des deux autres.
Num
Identité d'antisymétrisation
1
γ
[
μ
γ
ν
]
=
1
2
(
γ
μ
γ
ν
−
γ
ν
γ
μ
)
{\displaystyle \displaystyle \gamma ^{[\mu }\gamma ^{\nu ]}={\frac {1}{2}}(\gamma ^{\mu }\gamma ^{\nu }-\gamma ^{\nu }\gamma ^{\mu })}
2
γ
[
μ
γ
ν
γ
ρ
]
=
1
6
(
γ
μ
γ
ν
γ
ρ
+
γ
ρ
γ
μ
γ
ν
+
γ
ν
γ
ρ
γ
μ
−
γ
μ
γ
ρ
γ
ν
−
γ
ν
γ
μ
γ
ρ
−
γ
ρ
γ
ν
γ
μ
)
{\displaystyle \displaystyle \gamma ^{[\mu }\gamma ^{\nu }\gamma ^{\rho ]}={\frac {1}{6}}(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }+\gamma ^{\rho }\gamma ^{\mu }\gamma ^{\nu }+\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\mu }-\gamma ^{\mu }\gamma ^{\rho }\gamma ^{\nu }-\gamma ^{\nu }\gamma ^{\mu }\gamma ^{\rho }-\gamma ^{\rho }\gamma ^{\nu }\gamma ^{\mu })}
3
γ
[
μ
|
γ
ν
γ
|
ρ
]
=
1
2
(
γ
μ
γ
ν
γ
ρ
−
γ
ρ
γ
ν
γ
μ
)
{\displaystyle \displaystyle \gamma ^{[\mu |}\gamma ^{\nu }\gamma ^{|\rho ]}={\frac {1}{2}}(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }-\gamma ^{\rho }\gamma ^{\nu }\gamma ^{\mu })}
Remarques
L'identité no 1 définit tout simplement la notion de crochets d'antisymétrisation complète.
L'identité no 2 recèle en positif les permutations paires et en négatif les permutations impaires.
L'identité no 3 illustre la notion de crochets d'antisymétrisation partielle.
Les identités de trace sont particulièrement utiles lors de la résolution des diagrammes de Feynman en physique des particules.
Pour prouver ces identités on fera appel à trois propriétés de l'opérateur trace :
Additivité : tr(A + B ) = tr(A ) + tr(B )
Linéarité : tr(rA ) = r tr(A )
Cyclicité : tr(ABC ) = tr(CAB ) = tr(BCA )
ainsi qu'à celles du symbole de Levi-Civita (voir le point "extension à la matrice
γ
5
{\displaystyle \gamma ^{5}\,}
" ).
Num
Identité de trace
1
tr
(
γ
μ
)
=
0
{\displaystyle \operatorname {tr} (\gamma ^{\mu })=0}
2
La trace de tout produit d'un nombre impair de
γ
μ
{\displaystyle \gamma ^{\mu }}
est nulle.
3
La trace du produit de
γ
5
{\displaystyle \gamma ^{5}}
par un nombre impair de
γ
μ
{\displaystyle \gamma ^{\mu }}
est aussi nulle.
4
tr
(
γ
μ
γ
ν
)
=
4
g
μ
ν
{\displaystyle \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu })=4g^{\mu \nu }}
5
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
4
(
g
μ
ν
g
ρ
σ
−
g
μ
ρ
g
ν
σ
+
g
μ
σ
g
ν
ρ
)
{\displaystyle \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })=4(g^{\mu \nu }g^{\rho \sigma }-g^{\mu \rho }g^{\nu \sigma }+g^{\mu \sigma }g^{\nu \rho })}
6
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
γ
δ
γ
λ
)
=
4
g
μ
ν
(
g
ρ
σ
g
δ
λ
−
g
ρ
δ
g
σ
λ
+
g
ρ
λ
g
σ
δ
)
−
4
g
μ
ρ
(
g
ν
σ
g
δ
λ
−
g
ν
δ
g
σ
λ
+
g
ν
λ
g
σ
δ
)
+
4
g
μ
σ
(
g
ν
ρ
g
δ
λ
−
g
ν
δ
g
ρ
λ
+
g
ν
λ
g
ρ
δ
)
−
4
g
μ
δ
(
g
ν
ρ
g
σ
λ
−
g
ν
σ
g
ρ
λ
+
g
ν
λ
g
ρ
σ
)
−
4
g
μ
λ
(
g
ν
ρ
g
σ
δ
−
g
ν
σ
g
ρ
δ
+
g
ν
δ
g
ρ
σ
)
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda })=&4g^{\mu \nu }\left(g^{\rho \sigma }g^{\delta \lambda }-g^{\rho \delta }g^{\sigma \lambda }+g^{\rho \lambda }g^{\sigma \delta }\right)-\,4g^{\mu \rho }\left(g^{\nu \sigma }g^{\delta \lambda }-g^{\nu \delta }g^{\sigma \lambda }+g^{\nu \lambda }g^{\sigma \delta }\right)\\&+4g^{\mu \sigma }\left(g^{\nu \rho }g^{\delta \lambda }-g^{\nu \delta }g^{\rho \lambda }+g^{\nu \lambda }g^{\rho \delta }\right)-4g^{\mu \delta }\left(g^{\nu \rho }g^{\sigma \lambda }-g^{\nu \sigma }g^{\rho \lambda }+g^{\nu \lambda }g^{\rho \sigma }\right)\\&-4g^{\mu \lambda }\left(g^{\nu \rho }g^{\sigma \delta }-g^{\nu \sigma }g^{\rho \delta }+g^{\nu \delta }g^{\rho \sigma }\right)\end{aligned}}}
7
tr
(
γ
5
)
=
0
{\displaystyle \operatorname {tr} (\gamma ^{5})=0}
8
tr
(
γ
μ
γ
ν
γ
5
)
=
tr
(
γ
5
γ
μ
γ
ν
)
=
0
{\displaystyle \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{5})=\operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu })=0}
9
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
γ
5
)
=
−
4
i
ϵ
μ
ν
ρ
σ
{\displaystyle \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{5})=-4\mathrm {i} \epsilon ^{\mu \nu \rho \sigma }}
10
tr
(
γ
μ
1
…
γ
μ
n
)
=
tr
(
γ
μ
n
…
γ
μ
1
)
{\displaystyle \operatorname {tr} (\gamma ^{\mu 1}\dots \gamma ^{\mu n})=\operatorname {tr} (\gamma ^{\mu n}\dots \gamma ^{\mu 1})}
Démonstration
Preuve de l'identité no 1
On peut soit étudier les matrices directement, soit remarquer que :
tr
(
γ
ν
)
=
tr
(
γ
ν
γ
5
γ
5
)
.
{\displaystyle \operatorname {tr} (\gamma ^{\nu })=\operatorname {tr} (\gamma ^{\nu }\gamma ^{5}\gamma ^{5})\,.}
(en effet
(
γ
5
)
2
=
I
4
{\displaystyle (\gamma ^{5})^{2}=I_{4}}
)
En utilisant la propriété de cyclicité de la trace il vient :
tr
(
γ
ν
γ
5
γ
5
)
=
tr
(
γ
5
γ
ν
γ
5
)
{\displaystyle \operatorname {tr} (\gamma ^{\nu }\gamma ^{5}\gamma ^{5})=\operatorname {tr} (\gamma ^{5}\gamma ^{\nu }\gamma ^{5})}
et de par la relation d'anticommutation
{
γ
ν
,
γ
5
}
=
0
{\displaystyle \{\gamma ^{\nu },\gamma ^{5}\}=0}
on obtient
tr
(
γ
5
γ
ν
γ
5
)
=
−
tr
(
γ
ν
γ
5
γ
5
)
.
{\displaystyle \operatorname {tr} (\gamma ^{5}\gamma ^{\nu }\gamma ^{5})=-\operatorname {tr} (\gamma ^{\nu }\gamma ^{5}\gamma ^{5}).}
Comparer ces deux égalités permet de conclure.
Preuve de l'identité no 2
Il faut d'abord remarquer que
tr
(
γ
μ
)
=
0
.
{\displaystyle \operatorname {tr} (\gamma ^{\mu })=0\,.}
On rappelle également que la matrice
γ
5
{\displaystyle \gamma ^{5}\,}
vérifie :
(
γ
5
)
2
=
I
4
e
t
γ
μ
γ
5
=
−
γ
5
γ
μ
.
{\displaystyle \left(\gamma ^{5}\right)^{2}=I_{4}\quad \mathrm {et} \quad \gamma ^{\mu }\gamma ^{5}=-\gamma ^{5}\gamma ^{\mu }\,.}
Ainsi :
tr
(
γ
μ
γ
ν
γ
ρ
)
=
tr
(
γ
μ
γ
ν
γ
ρ
γ
5
γ
5
)
=
tr
(
γ
5
γ
μ
γ
ν
γ
ρ
γ
5
)
=
−
tr
(
γ
μ
γ
5
γ
ν
γ
ρ
γ
5
)
=
tr
(
γ
μ
γ
ν
γ
5
γ
ρ
γ
5
)
=
−
tr
(
γ
μ
γ
ν
γ
ρ
γ
5
γ
5
)
=
−
tr
(
γ
μ
γ
ν
γ
ρ
)
.
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho })&=\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{5}\gamma ^{5})\\&=\operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{5})\\&=-\operatorname {tr} (\gamma ^{\mu }\gamma ^{5}\gamma ^{\nu }\gamma ^{\rho }\gamma ^{5})\\&=\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{5}\gamma ^{\rho }\gamma ^{5})\\&=-\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{5}\gamma ^{5})\\&=-\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }).\end{aligned}}}
Ceci ne peut être satisfait que si
tr
(
γ
μ
γ
ν
γ
ρ
)
=
0
.
{\displaystyle \operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\right)=0\,.}
On voit bien qu'il faut un nombre impair d'anticommutations pour amener et conserver un signe moins. Cette propriété est vraie de manière générale quand un nombre impair de matrices gamma est impliqué.
C.Q.F.D.
Preuve de l'identité no 3
Si dans l'argument de la trace, un nombre impair de matrices gamma apparaît suivi par une matrice
γ
5
{\displaystyle \gamma ^{5}}
, on utilisera la cyclicité pour amener
γ
5
{\displaystyle \gamma ^{5}}
de la droite vers la gauche. Ceci laissera la trace invariante au signe près, chaque utilisation d'une anticommutation amenant un signe opposé au précédent. Cela signifie donc qu'en effectuant un nombre impair d'anticommutations, le signe sera négatif. Une trace égale à son opposée doit être nulle.
C.Q.F.D.
Preuve de l'identité no 4
Par cyclicité de la trace
1
2
tr
(
γ
μ
γ
ν
)
=
1
2
tr
(
γ
ν
γ
μ
)
.
{\displaystyle {\frac {1}{2}}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu })={\frac {1}{2}}\operatorname {tr} (\gamma ^{\nu }\gamma ^{\mu })\,.}
Ensuite, on a :
1
2
tr
(
γ
μ
γ
ν
)
+
1
2
tr
(
γ
ν
γ
μ
)
=
1
2
(
tr
(
γ
μ
γ
ν
)
+
tr
(
γ
ν
γ
μ
)
)
=
1
2
tr
(
γ
μ
γ
ν
+
γ
ν
γ
μ
)
=
1
2
tr
(
{
γ
μ
,
γ
ν
}
)
=
1
2
2
g
μ
ν
tr
(
I
4
)
=
4
g
μ
ν
.
{\displaystyle {\begin{aligned}{\frac {1}{2}}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu })+{\frac {1}{2}}\operatorname {tr} (\gamma ^{\nu }\gamma ^{\mu })&={\frac {1}{2}}(\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu })+\operatorname {tr} (\gamma ^{\nu }\gamma ^{\mu }))\\&={\frac {1}{2}}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }+\gamma ^{\nu }\gamma ^{\mu })\\&={\frac {1}{2}}\operatorname {tr} \left(\{\gamma ^{\mu },\gamma ^{\nu }\}\right)\\&={\frac {1}{2}}2g^{\mu \nu }\operatorname {tr} (I_{4})=4g^{\mu \nu }\,.\end{aligned}}}
C.Q.F.D.
Preuve de l'identité no 5
En utilisant la relation d'anticommutation on peut écrire
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
tr
(
γ
μ
γ
ν
(
2
g
ρ
σ
−
γ
σ
γ
ρ
)
)
=
2
g
ρ
σ
tr
(
γ
μ
γ
ν
)
−
tr
(
γ
μ
γ
ν
γ
σ
γ
ρ
)
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })&=\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\nu }(2g^{\rho \sigma }-\gamma ^{\sigma }\gamma ^{\rho })\right)\\&=2g^{\rho \sigma }\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\nu }\right)-\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\rho }\right)\end{aligned}}}
Dans le tout dernier terme on constate au passage que cette transformation a eu pour effet de faire se déplacer la matrice
γ
σ
{\displaystyle \gamma ^{\sigma }\,}
d'un cran de la droite vers la gauche. On poursuit :
tr
(
γ
μ
γ
ν
γ
σ
γ
ρ
)
=
tr
(
γ
μ
(
2
g
ν
σ
−
γ
σ
γ
ν
)
γ
ρ
)
=
2
g
ν
σ
tr
(
γ
μ
γ
ρ
)
−
tr
(
γ
μ
γ
σ
γ
ν
γ
ρ
)
{\displaystyle {\begin{aligned}\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\rho }\right)&=\operatorname {tr} \left(\gamma ^{\mu }(2g^{\nu \sigma }-\gamma ^{\sigma }\gamma ^{\nu })\gamma ^{\rho }\right)\\&=2g^{\nu \sigma }\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\rho }\right)-\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\sigma }\gamma ^{\nu }\gamma ^{\rho }\right)\end{aligned}}}
De même :
tr
(
γ
μ
γ
σ
γ
ν
γ
ρ
)
=
tr
(
(
2
g
μ
σ
−
γ
σ
γ
μ
)
γ
ν
γ
ρ
)
=
2
g
μ
σ
tr
(
γ
ν
γ
ρ
)
−
tr
(
γ
σ
γ
μ
γ
ν
γ
ρ
)
{\displaystyle {\begin{aligned}\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\sigma }\gamma ^{\nu }\gamma ^{\rho }\right)&=\operatorname {tr} \left((2g^{\mu \sigma }-\gamma ^{\sigma }\gamma ^{\mu })\gamma ^{\nu }\gamma ^{\rho }\right)\\&=2g^{\mu \sigma }\operatorname {tr} \left(\gamma ^{\nu }\gamma ^{\rho }\right)-\operatorname {tr} \left(\gamma ^{\sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\right)\end{aligned}}}
L'utilisation de l'identité
tr
(
γ
μ
γ
ν
)
=
4
g
μ
ν
{\displaystyle \operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\nu }\right)=4g^{\mu \nu }}
permet la simplification suivante :
2
g
ρ
σ
tr
(
γ
μ
γ
ν
)
=
2
g
ρ
σ
(
4
g
μ
ν
)
=
8
g
ρ
σ
g
μ
ν
.
{\displaystyle 2g^{\rho \sigma }\operatorname {tr} \left(\gamma ^{\mu }\gamma ^{\nu }\right)=2g^{\rho \sigma }(4g^{\mu \nu })=8g^{\rho \sigma }g^{\mu \nu }.\,}
De sorte que si on utilise les trois dernières identités, on obtient
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
8
g
ρ
σ
g
μ
ν
−
8
g
ν
σ
g
μ
ρ
+
8
g
μ
σ
g
ν
ρ
−
tr
(
γ
σ
γ
μ
γ
ν
γ
ρ
)
.
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })&=8g^{\rho \sigma }g^{\mu \nu }-8g^{\nu \sigma }g^{\mu \rho }+8g^{\mu \sigma }g^{\nu \rho }\\&-\ \operatorname {tr} \left(\gamma ^{\sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\right)\,.\end{aligned}}}
En utilisant la propriété de cyclicité de la trace il vient :
tr
(
γ
σ
γ
μ
γ
ν
γ
ρ
)
=
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
)
.
{\displaystyle \operatorname {tr} \left(\gamma ^{\sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\right)=\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })\,.}
Finalement,
2
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
8
g
ρ
σ
g
μ
ν
−
8
g
ν
σ
g
μ
ρ
+
8
g
μ
σ
g
ν
ρ
{\displaystyle 2\ \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })=8g^{\rho \sigma }g^{\mu \nu }-8g^{\nu \sigma }g^{\mu \rho }+8g^{\mu \sigma }g^{\nu \rho }\,}
ou
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
4
(
g
ρ
σ
g
μ
ν
−
g
ν
σ
g
μ
ρ
+
g
μ
σ
g
ν
ρ
)
.
{\displaystyle \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })=4\left(g^{\rho \sigma }g^{\mu \nu }-g^{\nu \sigma }g^{\mu \rho }+g^{\mu \sigma }g^{\nu \rho }\right)\,.}
C.Q.F.D.
Preuve de l'identité no 6
Pour démontrer que
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
γ
δ
γ
λ
)
=
4
g
μ
ν
(
g
ρ
σ
g
δ
λ
−
g
ρ
δ
g
σ
λ
+
g
ρ
λ
g
σ
δ
)
−
4
g
μ
ρ
(
g
ν
σ
g
δ
λ
−
g
ν
δ
g
σ
λ
+
g
ν
λ
g
σ
δ
)
+
4
g
μ
σ
(
g
ν
ρ
g
δ
λ
−
g
ν
δ
g
ρ
λ
+
g
ν
λ
g
ρ
δ
)
−
4
g
μ
δ
(
g
ν
ρ
g
σ
λ
−
g
ν
σ
g
ρ
λ
+
g
ν
λ
g
ρ
σ
)
−
4
g
μ
λ
(
g
ν
ρ
g
σ
δ
−
g
ν
σ
g
ρ
δ
+
g
ν
δ
g
ρ
σ
)
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda })=&4g^{\mu \nu }\left(g^{\rho \sigma }g^{\delta \lambda }-g^{\rho \delta }g^{\sigma \lambda }+g^{\rho \lambda }g^{\sigma \delta }\right)\\&-\,4g^{\mu \rho }\left(g^{\nu \sigma }g^{\delta \lambda }-g^{\nu \delta }g^{\sigma \lambda }+g^{\nu \lambda }g^{\sigma \delta }\right)\\&+4g^{\mu \sigma }\left(g^{\nu \rho }g^{\delta \lambda }-g^{\nu \delta }g^{\rho \lambda }+g^{\nu \lambda }g^{\rho \delta }\right)\\&-4g^{\mu \delta }\left(g^{\nu \rho }g^{\sigma \lambda }-g^{\nu \sigma }g^{\rho \lambda }+g^{\nu \lambda }g^{\rho \sigma }\right)\\&-4g^{\mu \lambda }\left(g^{\nu \rho }g^{\sigma \delta }-g^{\nu \sigma }g^{\rho \delta }+g^{\nu \delta }g^{\rho \sigma }\right)\end{aligned}}}
On emboîte d'abord différentes relations d'anticommutation du type :
{
γ
μ
,
γ
ν
}
=
2
g
μ
ν
.
{\displaystyle \{\gamma ^{\mu },\gamma ^{\nu }\}=2g^{\mu \nu }\,.}
Il vient :
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
γ
δ
γ
λ
)
=
2
g
μ
ν
tr
(
γ
ρ
γ
σ
γ
δ
γ
λ
)
−
tr
γ
ν
γ
μ
γ
ρ
γ
σ
γ
δ
γ
λ
)
=
2
g
μ
ν
tr
(
γ
ρ
γ
σ
γ
δ
γ
λ
)
−
2
g
μ
ρ
tr
(
γ
ν
γ
σ
γ
δ
γ
λ
)
+
2
g
μ
σ
tr
(
γ
ν
γ
ρ
γ
δ
γ
λ
)
−
2
g
μ
δ
tr
(
γ
ν
γ
ρ
γ
σ
γ
λ
)
+
2
g
μ
λ
tr
γ
ν
γ
ρ
γ
σ
γ
δ
)
−
tr
(
γ
ν
γ
ρ
γ
σ
γ
δ
γ
λ
γ
μ
)
.
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda })&=2g^{\mu \nu }\operatorname {tr} (\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda })-\operatorname {tr} \gamma ^{\nu }\gamma ^{\mu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda })\\&=2g^{\mu \nu }\operatorname {tr} (\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda })-2g^{\mu \rho }\operatorname {tr} (\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda })\\&+2g^{\mu \sigma }\operatorname {tr} (\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\delta }\gamma ^{\lambda })-2g^{\mu \delta }\operatorname {tr} (\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\lambda })\\&+2g^{\mu \lambda }\operatorname {tr} \gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta })-\operatorname {tr} (\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\delta }\gamma ^{\lambda }\gamma ^{\mu })\,.\end{aligned}}}
Puisque la trace est cyclique et que
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
4
(
g
μ
ν
g
ρ
σ
−
g
μ
ρ
g
ν
σ
+
g
μ
σ
g
ν
ρ
)
.
{\displaystyle \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })=4\left(g^{\mu \nu }g^{\rho \sigma }-g^{\mu \rho }g^{\nu \sigma }+g^{\mu \sigma }g^{\nu \rho }\right)\,.}
C.Q.F.D.
Preuve de l'identité no 7
On peut vérifier directement, ou l'observer par :
tr
(
γ
5
)
=
tr
(
γ
0
γ
0
γ
5
)
=
−
tr
(
γ
0
γ
5
γ
0
)
=
−
tr
(
γ
0
γ
0
γ
5
)
=
−
tr
(
γ
5
)
.
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{5})=\operatorname {tr} (\gamma ^{0}\gamma ^{0}\gamma ^{5})\\&=-\operatorname {tr} (\gamma ^{0}\gamma ^{5}\gamma ^{0})\\&=-\operatorname {tr} (\gamma ^{0}\gamma ^{0}\gamma ^{5})\\&=-\operatorname {tr} (\gamma ^{5}).\end{aligned}}}
Enfin additionnons
tr
(
γ
5
)
{\displaystyle \operatorname {tr} (\gamma ^{5})}
aux deux membres de l'égalité ci-dessus
2
tr
(
γ
5
)
=
0
.
{\displaystyle 2\operatorname {tr} (\gamma ^{5})=0\,.}
C.Q.F.D.
Preuve de l'identité no 8
Par cyclicité de l'opérateur trace, on a
tr
(
γ
μ
γ
ν
γ
5
)
=
tr
(
γ
5
γ
μ
γ
ν
)
.
{\displaystyle \operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{5})=\operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu })\,.}
Reste alors à démontrer que
tr
(
γ
5
γ
μ
γ
ν
)
=
0
.
{\displaystyle \operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu })=0\,.}
On considère les deux cas :
μ
n
e
q
ν
{\displaystyle \mu _{n}eq\nu \,}
et
μ
=
ν
{\displaystyle \mu =\nu \,}
.
cas 1 :
μ
≠
ν
{\displaystyle \mu \neq \nu \,}
On va ici utiliser le symbole de Levi-Civita.
Partons de
γ
5
=
i
γ
0
γ
1
γ
2
γ
3
{\displaystyle \gamma ^{5}=\mathrm {i} \gamma ^{0}\gamma ^{1}\gamma ^{2}\gamma ^{3}}
.
En utilisant le symbole de Levi-Civita, on a :
γ
5
=
−
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
ν
γ
σ
γ
ρ
{\displaystyle \gamma ^{5}=-\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\rho }}
(sans sommations)
où
ϵ
μ
ν
σ
ρ
=
1
{\displaystyle \epsilon _{\mu \nu \sigma \rho }=1}
si
μ
≠
ν
≠
σ
≠
ρ
{\displaystyle \mu \neq \nu \neq \sigma \neq \rho \,}
, et vaut 0 sinon.
On multiplie par le produit de deux matrices
γ
μ
γ
ν
{\displaystyle \gamma ^{\mu }\gamma ^{\nu }\,}
, avec
μ
{\displaystyle \mu \,}
différent de
ν
{\displaystyle \nu \,}
. Il vient :
γ
5
γ
μ
γ
ν
=
−
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
ν
γ
σ
γ
ρ
γ
μ
γ
ν
.
{\displaystyle \gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }=-\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\mu }\gamma ^{\nu }\,.}
En permutant cinq fois un couple de matrices gamma, les indices de ces matrices définissant à chaque fois un couple différent (en caractères gras ci-dessous), on récupère un signe plus.
En effet :
γ
5
γ
μ
γ
ν
=
−
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
ν
γ
σ
γ
ρ
γ
μ
γ
ν
=
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
ν
γ
σ
γ
μ
γ
ρ
γ
ν
=
−
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
ν
γ
σ
γ
μ
γ
ν
γ
ρ
=
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
ν
γ
μ
γ
σ
γ
ν
γ
ρ
=
−
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
ν
γ
μ
γ
ν
γ
σ
γ
ρ
=
i
ϵ
μ
ν
σ
ρ
γ
μ
γ
μ
γ
ν
γ
ν
γ
σ
γ
ρ
{\displaystyle {\begin{aligned}\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }&=-\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\mu }\gamma ^{\nu }\\&=\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\sigma }{\boldsymbol {\gamma ^{\mu }\gamma ^{\rho }}}\gamma ^{\nu }\\&=-\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\mu }{\boldsymbol {\gamma ^{\nu }\gamma ^{\rho }}}\\&=\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }\gamma ^{\nu }{\boldsymbol {\gamma ^{\mu }\gamma ^{\sigma }}}\gamma ^{\nu }\gamma ^{\rho }\\&=-\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\mu }{\boldsymbol {\gamma ^{\nu }\gamma ^{\sigma }}}\gamma ^{\rho }\\&=\mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\mu }{\boldsymbol {\gamma ^{\mu }\gamma ^{\nu }}}\gamma ^{\nu }\gamma ^{\sigma }\gamma ^{\rho }\end{aligned}}}
Examinons de plus près les valeurs des
μ
{\displaystyle \mu \,}
et
ν
{\displaystyle \nu \,}
.
On a
d'une part
μ
≠
ν
{\displaystyle \mu \neq \nu \,}
et d'autre part
μ
=
0
,
1
,
2
o
u
3
e
t
ν
=
0
,
1
,
2
o
u
3
:
{\displaystyle \mu \,=0,1,2\,ou\,3\,\quad \mathrm {et} \quad \nu \,=0,1,2\,ou\,3\,:}
Dans
(
1
)
{\displaystyle (1)\,}
on peut dès lors avoir :
(
γ
0
)
2
=
I
4
e
t
/
o
u
(
γ
i
)
2
=
−
I
4
a
v
e
c
i
=
1
,
2
,
3.
{\displaystyle \left(\gamma ^{0}\right)^{2}=I_{4}\quad \mathrm {et/ou} \quad \left(\gamma ^{i}\right)^{2}=-I_{4}\,\quad \mathrm {avec} \quad \,i=1,\,2,\,3.}
Donc :
γ
5
γ
μ
γ
ν
=
±
i
ϵ
μ
ν
σ
ρ
γ
σ
γ
ρ
.
{\displaystyle \gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }=\pm \mathrm {i} \epsilon _{\mu \nu \sigma \rho }\gamma ^{\sigma }\gamma ^{\rho }\,.}
Et on sait que
tr
(
γ
σ
γ
ρ
)
=
0
q
u
a
n
d
σ
≠
ρ
{\displaystyle \operatorname {tr} (\gamma ^{\sigma }\gamma ^{\rho })=0\quad \mathrm {quand} \quad \sigma \neq \rho \,}
ce qui est le cas ici.
Donc
tr
(
γ
5
γ
μ
γ
ν
)
=
0.
{\displaystyle \operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu })=0.}
cas 2 :
μ
=
ν
{\displaystyle \mu \,=\nu \,}
En ce cas
ϵ
μ
ν
σ
ρ
=
0
{\displaystyle \epsilon _{\mu \nu \sigma \rho }=0}
et donc la trace ne peut qu'être nulle.
C.Q.F.D.
Preuve de l'identité no 9
On part de
γ
5
=
i
γ
0
γ
1
γ
2
γ
3
.
{\displaystyle \gamma ^{5}=\mathrm {i} \gamma ^{0}\gamma ^{1}\gamma ^{2}\gamma ^{3}\,.}
On utilise d'abord l'égalité :
γ
5
=
i
γ
0
γ
1
γ
2
γ
3
=
−
i
ϵ
μ
ν
ρ
σ
γ
μ
γ
ν
γ
ρ
γ
σ
.
{\displaystyle \gamma ^{5}=\mathrm {i} \gamma ^{0}\gamma ^{1}\gamma ^{2}\gamma ^{3}=-\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\,.}
(sans sommations)
Ensuite on multiplie les deux membres extrêmes par
γ
μ
γ
ν
γ
ρ
γ
σ
{\displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }}
et on prend la trace :
tr
(
γ
5
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
tr
(
−
i
ϵ
μ
ν
ρ
σ
γ
μ
γ
ν
γ
ρ
γ
σ
γ
μ
γ
ν
γ
ρ
γ
σ
)
.
(
1
)
{\displaystyle \operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })=\operatorname {tr} (-\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })\,.\quad (1)}
(sans sommations)
On remarque que par permutations successives :
tr
(
γ
5
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
−
tr
(
γ
5
γ
μ
γ
ν
γ
σ
γ
ρ
)
=
tr
(
γ
5
γ
μ
γ
σ
γ
ν
γ
ρ
)
=
−
tr
(
γ
5
γ
σ
γ
μ
γ
ν
γ
ρ
)
=
tr
(
γ
5
γ
σ
γ
ν
γ
μ
γ
ρ
)
=
−
tr
(
γ
5
γ
σ
γ
ν
γ
ρ
γ
μ
)
=
tr
(
γ
5
γ
σ
γ
ρ
γ
ν
γ
μ
)
.
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })&=-\operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }{\boldsymbol {\gamma ^{\sigma }\gamma ^{\rho }}})\\&=\operatorname {tr} (\gamma ^{5}\gamma ^{\mu }{\boldsymbol {\gamma ^{\sigma }\gamma ^{\nu }}}\gamma ^{\rho })\\&=-\operatorname {tr} (\gamma ^{5}{\boldsymbol {\gamma ^{\sigma }\gamma ^{\mu }}}\gamma ^{\nu }\gamma ^{\rho })\\&=\operatorname {tr} (\gamma ^{5}\gamma ^{\sigma }{\boldsymbol {\gamma ^{\nu }\gamma ^{\mu }}}\gamma ^{\rho })\\&=-\operatorname {tr} (\gamma ^{5}\gamma ^{\sigma }\gamma ^{\nu }{\boldsymbol {\gamma ^{\rho }\gamma ^{\mu }}})\\&=\operatorname {tr} (\gamma ^{5}\gamma ^{\sigma }{\boldsymbol {\gamma ^{\rho }\gamma ^{\nu }}}\gamma ^{\mu }).\end{aligned}}}
Donc on peut effectuer un renversement complet de l'ordre des quatre facteurs (abcd devenant dcba ) afférents aux quatre dernières matrices gamma dans le membre de droite de
(
1
)
{\displaystyle (1)}
tr
(
γ
5
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
tr
(
−
i
ϵ
μ
ν
ρ
σ
γ
μ
γ
ν
γ
ρ
γ
σ
γ
σ
γ
ρ
γ
ν
γ
μ
)
.
{\displaystyle \operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })=\operatorname {tr} (-\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\nu }\gamma ^{\mu })\,.}
(sans sommations)
Après avoir mis les constantes pour l'opérateur trace en évidence, il est alors aisé de réaliser l'appariement des matrices gamma comme suit :
tr
(
γ
5
γ
μ
γ
ν
γ
ρ
γ
σ
)
=
−
i
ϵ
μ
ν
ρ
σ
tr
(
γ
μ
γ
ν
γ
ρ
γ
σ
γ
σ
γ
ρ
γ
ν
γ
μ
)
=
−
i
ϵ
μ
ν
ρ
σ
tr
(
(
γ
0
)
2
(
γ
1
)
2
(
γ
2
)
2
(
γ
3
)
2
)
=
−
i
ϵ
μ
ν
ρ
σ
tr
(
I
4
(
−
I
4
)
(
−
I
4
)
(
−
I
4
)
)
=
i
ϵ
μ
ν
ρ
σ
tr
(
I
4
)
=
4
i
ϵ
μ
ν
ρ
σ
=
−
4
i
ϵ
μ
ν
ρ
σ
{\displaystyle {\begin{aligned}\operatorname {tr} (\gamma ^{5}\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma })&=-\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\operatorname {tr} (\gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\nu }\gamma ^{\mu })\\&=-\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\operatorname {tr} (\left(\gamma ^{0}\right)^{2}\left(\gamma ^{1}\right)^{2}\left(\gamma ^{2}\right)^{2}\left(\gamma ^{3}\right)^{2})\\&=-\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\operatorname {tr} (I_{4}\left(-I_{4}\right)\left(-I_{4}\right)\left(-I_{4}\right))\\&=\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\operatorname {tr} (I_{4})\\&=4\mathrm {i} \epsilon _{\mu \nu \rho \sigma }\\&=-4\mathrm {i} \epsilon ^{\mu \nu \rho \sigma }\end{aligned}}}
C.Q.F.D.
Preuve de l'identité no 10
On note le produit de
n
{\displaystyle n}
matrices gamma par
Γ
=
γ
μ
1
γ
μ
2
…
γ
μ
n
.
{\displaystyle \Gamma =\gamma ^{\mu 1}\gamma ^{\mu 2}\dots \gamma ^{\mu n}.}
Son conjugué hermitien est alors :
Γ
†
=
γ
μ
n
†
…
γ
μ
2
†
γ
μ
1
†
=
γ
0
γ
μ
n
γ
0
…
γ
0
γ
μ
2
γ
0
γ
0
γ
μ
1
γ
0
=
γ
0
γ
μ
n
…
γ
μ
2
γ
μ
1
γ
0
{\displaystyle {\begin{aligned}\Gamma ^{\dagger }&=\gamma ^{\mu n\dagger }\dots \gamma ^{\mu 2\dagger }\gamma ^{\mu 1\dagger }\\&=\gamma ^{0}\gamma ^{\mu n}\gamma ^{0}\dots \gamma ^{0}\gamma ^{\mu 2}\gamma ^{0}\gamma ^{0}\gamma ^{\mu 1}\gamma ^{0}\\&=\gamma ^{0}\gamma ^{\mu n}\dots \gamma ^{\mu 2}\gamma ^{\mu 1}\gamma ^{0}\end{aligned}}}
(toutes les
γ
0
{\displaystyle \gamma ^{0}}
se simplifient sauf la première et la dernière car
(
γ
0
)
2
=
I
4
{\displaystyle (\gamma ^{0})^{2}=I_{4}}
)
Si l'on conjugue encore une fois avec
γ
0
{\displaystyle \gamma ^{0}}
de manière à se débarrasser des deux
γ
0
{\displaystyle \gamma ^{0}}
restantes, on voit que
γ
0
Γ
†
γ
0
{\displaystyle \gamma ^{0}\Gamma ^{\dagger }\gamma ^{0}}
est l'inverse de
Γ
{\displaystyle \Gamma }
. Maintenant, en utilisant les propriétés de la fonction trace, on a :
tr
(
γ
0
Γ
†
γ
0
)
=
tr
(
Γ
†
)
=
tr
(
Γ
∗
)
=
tr
(
Γ
)
{\displaystyle \operatorname {tr} (\gamma ^{0}\Gamma ^{\dagger }\gamma ^{0})=\operatorname {tr} (\Gamma ^{\dagger })=\operatorname {tr} (\Gamma ^{*})=\operatorname {tr} (\Gamma )}
C.Q.F.D.
Ces remarquables identités ont permis la mise au point d'algorithmes très puissants visant à simplifier les calculs de produits ou de traces impliquant un grand nombre de matrices gamma[ 2] .
On note ici
S
(
impair
)
=
γ
ν
λ
1
…
γ
ν
λ
2
k
+
1
{\displaystyle \displaystyle S^{({\textrm {impair}})}=\gamma ^{\nu \lambda _{1}}\dots \gamma ^{\nu \lambda _{2k+1}}}
, un produit d'un nombre impair de matrices gamma
S
(
pair
)
=
γ
ν
λ
1
…
γ
ν
λ
2
k
{\displaystyle \displaystyle S^{({\textrm {pair}})}=\gamma ^{\nu \lambda _{1}}\dots \gamma ^{\nu \lambda _{2k}}}
, un produit d'un nombre pair de matrices gamma
L'indice R signifie que l'on prend le produit de matrices gamma dans l'ordre inverse.
Ainsi :
S
R
(
impair
)
=
γ
ν
λ
2
k
+
1
γ
ν
λ
2
k
…
γ
ν
λ
1
{\displaystyle S_{R}^{({\textrm {impair}})}=\gamma ^{\nu \lambda _{2k+1}}\gamma ^{\nu \lambda _{2k}}\dots \gamma ^{\nu \lambda _{1}}}
S
R
(
pair
)
=
γ
ν
λ
2
k
…
γ
ν
λ
1
{\displaystyle S_{R}^{({\textrm {pair}})}=\gamma ^{\nu \lambda _{2k}}\dots \gamma ^{\nu \lambda _{1}}}
Le prime dans
S
R
′
(
i
m
p
a
i
r
)
{\displaystyle \displaystyle S_{R}^{'(impair)}}
est là pour attirer l'attention sur le fait que le nombre pair de matrices gamma est obtenu en multipliant un nombre de matrices impair par une seule matrice gamma de sorte que
S
R
(
pair
)
=
γ
λ
S
R
′
(
impair
)
.
{\displaystyle S_{R}^{({\textrm {pair}})}=\gamma ^{\lambda }S_{R}^{'({\textrm {impair}})}\,.}
Num
Identité de Chisholm
1
γ
μ
S
(
impair
)
γ
μ
=
−
2
S
R
(
impair
)
{\displaystyle \displaystyle \gamma ^{\mu }S^{({\textrm {impair}})}\gamma _{\mu }=-2S_{R}^{({\textrm {impair}})}}
2
γ
μ
S
(
p
a
i
r
)
γ
μ
=
γ
μ
γ
λ
S
R
′
(
impair
)
γ
μ
=
2
γ
λ
(
S
R
′
(
impair
)
)
+
2
(
S
R
′
(
impair
)
)
γ
λ
{\displaystyle \displaystyle \gamma ^{\mu }S^{(pair)}\gamma _{\mu }=\gamma ^{\mu }\gamma ^{\lambda }S_{R}^{'({\textrm {impair}})}\gamma _{\mu }=2\gamma ^{\lambda }(S_{R}^{'({\textrm {impair}})})+2(S_{R}^{'({\textrm {impair}})})\gamma ^{\lambda }}
3
γ
μ
tr
(
γ
μ
S
(
i
m
p
a
i
r
)
)
=
2
(
S
(
impair
)
+
S
R
(
impair
)
)
{\displaystyle \displaystyle \gamma _{\mu }\operatorname {tr} (\gamma _{\mu }S^{(impair)})=2(S^{({\textrm {impair}})}+S_{R}^{({\textrm {impair}})})}
À propos de l'identité
no 1
Si l'on y regarde de plus près on constate que l'identité :
γ
μ
γ
ν
γ
ρ
γ
σ
γ
μ
=
−
2
γ
σ
γ
ρ
γ
ν
{\displaystyle \displaystyle \gamma ^{\mu }\gamma ^{\nu }\gamma ^{\rho }\gamma ^{\sigma }\gamma _{\mu }=-2\gamma ^{\sigma }\gamma ^{\rho }\gamma ^{\nu }}
vue dans la section Identités de contraction , n'est qu'un cas particulier de cette première identité de Chisholm.
(en) Mark Srednicki, Quantum Field Theory , Cambridge University Press, 2007 (ISBN 978-0-521-86449-7 ) Voir chapitre 47 .
(en) Michio Kaku, Quantum Field Theory (ISBN 9780195091588 ) , appendix A
(en) A. Zee, Quantum Field Theory in a Nutshell 2e ed. , Princeton, New Jersey, Princeton University Press, 2010 (ISBN 978-0-691-14034-6 ) .
(en) M. Peskin et D. Schroeder, An Introduction to Quantum Field Theory , Westview Press, 1995 (ISBN 0-201-50397-2 ) Voir chapitre 3.2 .
(en) Claude Itzykson et Jean-Bernard Zuber, Quantum Field Theory , 1980 (ISBN 0-07-066353-X ) .
Jean Hladik, Le calcul tensoriel en physique , 1993), masson, paris (ISBN 2-225-84144-6 )
(en) Yves R. Talpaert, Tensor Analysis and Continuum Mechanics , 2002 (ISBN 1-4020-1055-9 )
Paul Adrien Maurice Dirac, Les principes de la mécanique quantique, traduction de la 4e éd. , Lausanne, Presses polytechniques et universitaires romandes, 2009 (ISBN 978-2-88074-800-5 )
W. Pauli, « Contributions mathématiques à la théorie des matrices de Dirac », Ann. Inst. Henri Poincaré , vol. 6, 1936 , p. 109 (lire en ligne )
(en) W.A. Horowitz, Traces of Gamma Matrices , 17 novembre 2010
(en) J. B. Formiga, A list of identities made with products between two different generators of the Clifford algebra , Centro de Ciências de Natureza, Piaui, Brazil, 27 septembre 2012
Jean-Pierre Derendinger, Théorie quantique des champs , Lausanne, coll. « Presses polytechniques et universitaires romandes », 2001 (ISBN 2-88074-491-1 )
(en) V.I. Borodulin, R.N. Rogalyov et S.R. Slabospitsky, CORE, COmpendium of RElations , Protvino, Moscow, Russia, Institute for High Energy Physics, 1995