1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

A Lagrangian of a monopole (Einstein notation is used)

  1. May 4, 2016 #1
    Hi everyone,

    I am trying to calculate the equation of motion of a charged particle in the field of a monopole.

    The magnetic field of a monopole of strength g is given by:
    [tex] \textbf{B} = g \frac{\textbf{r}}{r^3} [/tex]
    And the Lagrangian by:
    [tex] \mathcal{L} = \frac{m\dot{\textbf{r}}^2}{2} + e\textbf{A}\cdot \dot{\textbf{r}}[/tex]
    Where e is the electric charge and A is the vector potential of the magnetic field [tex] \textbf{B} = \nabla \times \textbf{A} [/tex]
    The equations of motion (1) should become:
    [tex] m \ddot{\textbf{r}} = e(\dot{\textbf{r}} \times \textbf{B}) [/tex]

    The EL equation is:
    [tex] \frac{d}{dt}\frac{\partial\mathcal{L}}{\partial \dot{x_i}} = \frac{\partial\mathcal{L}}{\partial x_i}[/tex]

    In Einstein notation, I get the corresponding terms:
    [tex] \frac{d}{dt}\frac{\partial\mathcal{L}}{\partial \dot{x_i}} = m\ddot{x_i}\\
    \frac{\partial\mathcal{L}}{\partial x_i} = e \dot{x_j}\partial_i A_j[/tex]

    However, this does not correspond to the equations of motion given by equation (1) translated into an Einstein notation:
    [tex] m\ddot{x_i} = e \dot{x_j}\partial_i A_j - e\dot{x_j}\partial_j A_i [/tex]
    Since this equation includes the extra term:
    [tex]- e\dot{x_j}\partial_j A_i [/tex]
    Which corresponds to:
    [tex] -e(\dot{\textbf{r}} \cdot \nabla)\textbf{A} [/tex]

    I have the feeling everything is alright, but this extra term becomes zero since the vector potential only has components perpendicular to r, however, I cannot find a way to prove it mathematically.

    It is quite some work to type all the steps in between, and I tried to stay as clear as possible.
    If there are any questions or if something is not clear, I would gladly like to answer them.

    Thanks in advance,
  2. jcsd
  3. May 4, 2016 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    I'm not exactly sure what you're doing, but you can't describe a monopole using [itex]\vec{B} = \nabla \times \vec{A}[/itex]. That's because

    [itex]\nabla \cdot \vec{B} \propto \rho_m[/itex] the monopole density
    [itex]\nabla \cdot (\nabla \times \vec{A}) = 0[/itex]
  4. May 4, 2016 #3


    User Avatar
    Staff Emeritus
    Science Advisor

    But also,

    [itex]\frac{\partial \mathcal{L}}{\partial \dot{r}} = m \dot{r} + e \vec{A}[/itex]

    So [itex]\frac{d}{dt} \frac{\partial \mathcal{L}}{\partial \dot{r}} = m \ddot{r} + e \frac{d}{dt} \vec{A}[/itex], not just [itex]m \ddot{r}[/itex]
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted