1

I know that the interaction terms of the Lagrangian of electromagnetism are given by

$$L_{int}=-q\phi (\mathbf{x},t)+q\mathbf{v}(t)\cdot \mathbf{A}(\mathbf{x},t).$$

The above equation is replaced by terms involving a continuous charge density $\rho$ and current density $\mathbf{j}$. The resulting Lagrangian density for the electromagnetic field is: $$\mathcal{L}=-\rho \phi +\mathbf{j}\cdot \mathbf{A}+\frac{\epsilon _{0}}{2}E^{2}-\frac{1}{2\mu _{0}}B^{2} .$$

The first problem is that I know where the first two terms $-\rho \phi +\mathbf{j}\cdot \mathbf{A}$ come from but I don't know where the last two terms $\frac{\epsilon _{0}}{2}E^{2}-\frac{1}{2\mu _{0}}B^{2}$ come from.

Next varying the Lagrangian density with respect to $\phi$ and $\mathbf{A}$, we get Gauss' law $$0=-\rho +\epsilon _{0}\triangledown \cdot \mathbf{E}$$ and Ampère's law $$0=\mathbf{j}+\epsilon _{0}\frac{\partial \mathbf{E}}{\partial t}-\frac{1}{\mu _{0}}\triangledown \times \mathbf{B},$$ respectively. The second problem is that I don't know how to calculate these variations clearly.

  • Lagrangians are not derived. They are constructed such that they yield the correct equations of motion. 2. The second part of your question seems to just be how to derive the Euler-Lagrange equations, which is explained in many places.
  • – ACuriousMind Sep 22 '16 at 16:36
  • @ACuriousMind yes I know that Lagrangian is not derived, but my problem is how to transfer Lagrangian to Lagrangian density. – William Huang Sep 22 '16 at 16:39
  • Then what exactly do you mean by your "first problem"? – ACuriousMind Sep 22 '16 at 16:41
  • @ACuriousMind I know what the Euler-Lagrange equation is, but what bothers me is the technical detail – William Huang Sep 22 '16 at 16:45
  • Possible duplicates: http://physics.stackexchange.com/q/3005/2451 , https://physics.stackexchange.com/q/34241/2451 , http://physics.stackexchange.com/q/51169/2451 , http://physics.stackexchange.com/q/64272/2451 and links therein. – Qmechanic Sep 22 '16 at 18:14
  • Although late in the party, take a look in my answer here : Deriving Lagrangian density for electromagnetic field ( http://physics.stackexchange.com/questions/34241/deriving-lagrangian-density-for-electromagnetic-field/270950#270950 ). – Frobenius Jan 17 '17 at 19:05