1
$\begingroup$

Suppose $M$ be any smooth manifold, and $E$ be a hermitian line bundle over $M$ with hermitian metric $h$. The action functional $S(\phi, \psi)$ for two section variabled $\phi$ and $\psi$ is given by $$ S(\phi, \psi) = \int_M h(\phi, \psi)\, \mathrm{d}V_M $$ Now, I would like to find a field equations associated with $S$, and it's equivalent to find the local expression of the Euler-Lagrange equation on each charts.

Let $U$ be a chart of $M$, and take a trivialization $\Phi$ of $E|U \cong U \times \mathbb{C}$ such that $\Phi^* h$ is just a trivial complex sesquilinear form. Suppose $\varphi^\ast$ and $\varphi$ be a local representation of $\psi$ and $\phi$. Then, the local Euler-Lagrange equation associated to $S$ is equivalent to $$ \varphi^* = 0\\ \varphi = 0 $$ since the local Lagrangian is given by $\mathcal{L}|\Phi = \varphi^* \varphi$.

The action functional is too simple here and so the equations also become too simple, but in fact, this is just a toy model for checking whether my approach is correct or not.

Any comments, improvements and advices are appreciated.

EDIT In fact, the question is that how can we deal with the field equations on smooth manifold if the field is considered as the section of some hermitian line bundle.

I know that there is a similar question in MO, but I'm still wondering how to get such field equations from the globally defined action functional on the bundle, and solve them. Is it possible without any specific local expression for general case?

or any references about this subject is also appreciated. I think I'm too ignorant of the subject.

$\endgroup$
5
  • $\begingroup$ I don't understand. Where is the potential confusion? And why introduce charts and trivializations in the first place? $\endgroup$ Commented Dec 17, 2022 at 10:44
  • $\begingroup$ @IgorKhavkine Sorry for lack of rigour. I introduce charts and trivializations to get a Euler-Lagrange equation of given action functional, since the field is not a complex function on $\mathbb{R}^n$, but the section of arbitrary complex line bundle on some manifold $M$. I don't know much about this. But, I've seen this in some paper. Could I ask what is the potential confusion? I didn't intend something physical, but only mathematical. If it were possible, then it would be good, but it makes the problem a little sophisticated and so, I wouldn't check I goes to the right direction. $\endgroup$
    – ChoMedit
    Commented Dec 17, 2022 at 15:59
  • $\begingroup$ I don't think this is the right place for a "please check my work" request. Even then, you haven't indicated where your own confusion might be in this (unnecessary complicated by charts) calculation. $\endgroup$ Commented Dec 17, 2022 at 22:25
  • $\begingroup$ @IgorKhavkine I apologize for that.. I now clarify where my confusion is. In fact, why did you say that the chart is unnecessary complicated? Are there something good tools to get a field equations from action functional without trivializations? I mean, most of people doesn't have an attention on specific local expression of the field equation? $\endgroup$
    – ChoMedit
    Commented Dec 18, 2022 at 2:47
  • 2
    $\begingroup$ No charts needed to define the directional (or Fréchet) derivatives. There are too many references to pick just one canonical one, but see for example Classical Field Theory by Deligne & Freed from Quantum fields and strings: a course for mathematicians (AMS, 1999). $\endgroup$ Commented Dec 18, 2022 at 7:48

0

You must log in to answer this question.

Browse other questions tagged .