1
$\begingroup$

If I have a look at the most generalized version of the Dirac Lagrangian I can not identify all terms with there contrubution in the Feynman diagrams.

$$\mathcal{L} = \sum_a (\overline{\psi}_a {\not}{\partial} \psi^a-{\not}{\partial} \psi^a \overline{\psi}_a+{\psi}_a {\not}{\partial} \overline{\psi}^a-{\not}{\partial} \overline{\psi}^a {\psi}_a -m \overline{\psi}^a\psi_a +m\psi^a \overline{\psi}_a)$$

The last two terms should represent the creation of a fermion and an antifermion but which two kinetic terms correspond to which propagators?

$\endgroup$
4
  • $\begingroup$ THe last two terms are mass terms. $\endgroup$
    – Prahar
    Commented Feb 17, 2017 at 0:54
  • $\begingroup$ I guess that the first and the fourth correspond to the electron (fermion) and the second and third to the positron (antifermion)? $\endgroup$
    – Alpha001
    Commented Feb 17, 2017 at 1:13
  • $\begingroup$ Remark: there are several factors of $\frac12$ and $i$ missing in $\mathcal L$ (and some signs might be wrong, I haven't checked in detail). $\endgroup$ Commented Feb 17, 2017 at 23:37
  • $\begingroup$ @Alpha001 how do you known which term represents the fermion and which one the antifermion? $\endgroup$
    – Schiele
    Commented May 10, 2023 at 23:32

1 Answer 1

3
$\begingroup$

Integrating by parts, all the terms in $\mathcal L$ collapse into one, $$ \mathcal L=\bar\psi(i\not\partial+m)\psi+\text{boundary terms} $$

Therefore, in terms of Feynman diagrams, you have $$ ---\stackrel{p}{\blacktriangleright} --- \qquad=\qquad \frac{1}{\not p-m} $$

Your Lagrangian and the standard Dirac Lagrangian differ by a total derivative. Therefore, they give rise to the same action, same equations of motion, same dynamics, and same diagrammatics. Your "generalised" theory is just as general as the standard Dirac theory.


In principle, you could use the standard rules of perturbation theory to read off the Feynman rules of a total derivative in the Lagrangian; but the result is proportional to the sum of all momenta incoming to the vertex, and therefore such a term vanishes by conservation of momentum. The details are left to the reader.

$\endgroup$
2
  • $\begingroup$ Is this also true if the above Lagrangian consists of field operators? They would in general not commute. $\endgroup$
    – Alpha001
    Commented Feb 17, 2017 at 23:39
  • $\begingroup$ @Alpha001 yes, I was referring to operators in my answer. If you want, you can think about this in terms of path integrals, where the $\psi$'s are Grassmann variables, and therefore their commutators are trivial (they are anti-commuting numbers). $\endgroup$ Commented Feb 18, 2017 at 8:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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