4
$\begingroup$

Is the matrix $\sigma_{\mu\nu}$

$$\sigma_{\mu\nu} = \frac{i}{2} [\gamma_\mu, \gamma_\nu]$$

defined in FeynRules, FeynCalc or any similar packages? I know that $\gamma_\mu$ is Ga[mu] in FeynRules.

$\endgroup$
3
  • $\begingroup$ PS it won't let me enclose my equation in $$ for latex... $\endgroup$
    – innisfree
    Nov 13 '17 at 0:51
  • $\begingroup$ Not sure what you did wrong; enclosing in $$ works, as you might now see. $\endgroup$
    – J. M.'s torpor
    Nov 13 '17 at 3:23
  • $\begingroup$ Funny, it said that my question contained code that wasn't formatted properly and wouldn't let me post. The preview looked fine. $\endgroup$
    – innisfree
    Nov 13 '17 at 3:25
6
$\begingroup$

The $\gamma$ matrices are built-in, but undocumented, as Internal`DiracGammaMatrix[]. Their indexing is also a bit different from the wiki page:

Table[Internal`DiracGammaMatrix[k, "Basis" -> "Dirac"] // MatrixForm, {k, 4}]

$$\{\begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & -1 \\ \end{pmatrix}, \begin{pmatrix} 0 & 0 & 0 & 1 \\ 0 & 0 & 1 & 0 \\ 0 & -1 & 0 & 0 \\ -1 & 0 & 0 & 0 \\ \end{pmatrix}, \begin{pmatrix} 0 & 0 & 0 & -i \\ 0 & 0 & i & 0 \\ 0 & i & 0 & 0 \\ -i & 0 & 0 & 0 \\ \end{pmatrix}, \begin{pmatrix} 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & -1 \\ -1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ \end{pmatrix}\}$$

(Other possible settings for "Basis" include "Chiral" (the default) and "Majorana".)

Thus, you can implement the commutator like this:

Options[diracCommutator] = Options[Internal`DiracGammaMatrix];
diracCommutator[p_, q_, opts : OptionsPattern[]] := 
   With[{pm = Internal`DiracGammaMatrix[p, opts], qm = Internal`DiracGammaMatrix[q, opts]},
        I (pm.qm - qm.pm)/2]

and then you can do e.g.

diracCommutator[2, 4, "Basis" -> "Dirac"] // MatrixForm

$$\begin{pmatrix} 0 & i & 0 & 0 \\ -i & 0 & 0 & 0 \\ 0 & 0 & 0 & i \\ 0 & 0 & -i & 0 \\ \end{pmatrix}$$

$\endgroup$
2
  • $\begingroup$ This is without using FeynCalc, right? $\endgroup$
    – innisfree
    Nov 13 '17 at 7:11
  • $\begingroup$ Nope, it's built-in, but undocumented. $\endgroup$
    – J. M.'s torpor
    Nov 13 '17 at 7:12
1
$\begingroup$

In FeynCalc there is DiracSigma

DiracSigma[GA[mu], GA[nu]]

with

?DiracSigma

DiracSigma[a, b] stands for I/2*(a . b - b . a) in 4 dimensions. a and b must have Head DiracGamma, DiracMatrix or DiracSlash. Only antisymmetry is implemented.

However, you cannot have explicit Dirac indices attached to the Dirac matrices. This is something planned for the future (which is needed to use FeynCalc with QGraf)

$\endgroup$
0
1
$\begingroup$

The matrix $\sigma_{\mu\nu}$ is given by Sig[mu, nu] in FeynRules, a related package, and defined

Sig[mu_,nu_,ss1_,ss2_]->I/2 TensDot[Ga[mu].Ga[nu]][ss1,ss2]-I/2TensDot[Ga[nu].Ga[mu]][ss1,ss2]},\[Infinity],Heads->True];

in Interfaces/FeynArtsInterface.m.

$\endgroup$
1
$\begingroup$

In Package-X, the object DiracS (σ) represents the commutator of Dirac gamma matrices.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.