0
$\begingroup$

In https://arxiv.org/abs/1011.1145 the following rather general gauge algebra identity (2.4) is used

$$\delta_{gct}(\xi) B_\mu^{\>\>A} + \xi^\lambda R_{\mu\lambda}^{\quad\! A} -\sum_{\{C\}}\delta(\xi^\lambda B_\lambda^{\>\>C})B_\mu^{\>\>A} = 0 $$

The free $A$ index references any chosen gauge transformation in the gauge algebra, $B_\mu^{\>\>A}$ its corresponding gauge field / connection, $R_{\mu\nu}^{\quad\! A}$ its corresponding field strength / curvature, and $\xi = \xi^\lambda \partial_\lambda$ the generator of the general coordinate transformations $\delta_{gct}(\xi)$. The last sum over $C$ is the sum over all gauge transformations $\delta_C$ in the gauge algebra.

What follows is clear (and is not relevant to my post) but the author provides no explanation of this identity nor cites a reference where to look it up. Presumably it is valid for any gauge algebra that closes off-shell? Anyway, this relation is not clear to me. Can anyone please provide a derivation of this identity or a reference where it is derived? Thank you.

$\endgroup$
0
0
$\begingroup$

OK, I figured it out myself.

$$\begin{eqnarray}\delta_{gct}(\xi) B_\mu^{\>\>A} &\;=\;& \mathcal L_\xi B_\mu^{\>\>A} \\ &=& \xi^\lambda\partial_\lambda B_\mu^{\>\>A} + B_\lambda^{\>\>A}\partial_\mu \xi^\lambda \\ &=& \xi^\lambda\partial_{[\lambda} B_{\mu]}^{\>\>A} + \partial_\mu(\xi^\lambda B_\lambda^{\>\>A})\\ &=& \xi^\lambda \big( \partial_{[\lambda} B_{\mu]}^{\>\>A} + f_{BC}^{\quad A} B_\lambda^{\>\>B}B_\mu^{\>\>C} \big) + \partial_\mu(\xi^\lambda B_\lambda^{\>\>A}) + f_{BC}^{\quad A} B_\mu^{\>\>B}(\xi^\lambda B_\lambda^{\>\>C})\\ &=& \xi^\lambda R_{\lambda\mu}^{\quad \!A} + \delta \big(\epsilon = {\textstyle\sum}_C \xi^\lambda B_\lambda^{\>\>C}) B_\mu^{\>\>A} \end{eqnarray}$$ Notation: Here $\mathcal L$ is the Lie derivative, index antisymetrization is normalized as $T_{[ab]} \equiv T_{ab}-T_{ba}$, and summation on a contracted internal gauge index $A, B, C,...$ means over the entire gauge algebra. So for gauged Poincare the sum on $B^C$ runs over the vielbein $e^a$ and Lorentz connection $\omega_a^{\>\,b}$. The last term happens to be the action of gauge algebra transformation $\delta(\epsilon)$ acting on the fixed connection $B_\mu^{\>\>A}$, but with the usual gauge parameter $\epsilon$ replaced by $\sum_C\xi^\lambda B_\lambda^{\>\>C}$.

$\endgroup$

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.