Consider a ralativistic particle of rest mass $m$ and electric charge $e $ moving in electromagnetic field with four-potential ${\displaystyle A^{\mu}=(\phi ,\mathrm {A} )} $ in vacuum, then the Hamilton–Jacobi equation has the form

$$g^{\mu \nu}\left ( \frac{\partial S}{\partial x^{\mu}} + \frac {e}{c}A_{\mu} \right ) \left ( \frac{\partial S}{\partial x^{\nu}} + \frac {e}{c}A_{\nu} \right ) = m^2 c^2\tag{1}$$

or more compact expressed as Minkowski product

$$ \left( \frac{\partial S}{\partial x^{\mu}} + \frac {e}{c}A_{\mu} \right ) \left ( \frac{\partial S}{\partial x_{\mu}} + \frac {e}{c}A^{\mu} \right ) = m^2 c^2 \tag{2}$$

here we denote $g^{\mu \nu}$ the metric tensor with signature $(+ - - -)$ and $S$ is the action function from Hamilton-Jacobi-theory.

Especially $S$ satisfy the equation

$$p_{\mu}= \nabla_{\mu}S := \frac{\partial S}{\partial x^{\mu}}\tag{3}$$

where $p_{\mu}$ is the four momentum and $\nabla_{\mu}$ the four gradient.

Now I have following two questions:

  1. Does anybody have a reference for a rigorous derivation for $$ \left( \frac{\partial S}{\partial x^{\mu}} + \frac {e}{c}A_{\mu} \right ) \left ( \frac{\partial S}{\partial x_{\mu}} + \frac {e}{c}A^{\mu} \right ) = m^2 c^2 .\tag{4}$$

  2. It is known that applying method of characteristics to the PDE

$$F(S,\frac{\partial S}{\partial x^{\mu}} ,x^{\mu}):= \left( \frac{\partial S}{\partial x^{\mu}} + \frac {e}{c}A_{\mu} \right ) \left ( \frac{\partial S}{\partial x_{\mu}} + \frac {e}{c}A^{\mu} \right ) - m^2 c^2 =0\tag{5}$$

one can derive the relative Lorentz equation

$${\displaystyle {\frac {\mathrm {d} p^{\mu }}{\mathrm {d} \tau }}=eF^{\mu \nu }p_{\nu }}\tag{6}$$

with electromagnetic tensor $$F^{\mu \nu }:= \frac{\partial A_{\mu}}{\partial x^{\nu}}- \frac{\partial A_{\nu}}{\partial x^{\mu}}\tag{7}$$ and four momentum $p_{\mu}$.

Here I'm also looking for an explicit derivation of LE from the HJE using characteristics.

Indeed, the method of characteristics transform a PDE into a system of ODE with respect parametrizing variable $\tau$:

$$\frac{\partial p_{\mu}}{\partial \tau}= -\frac{\partial F}{\partial x^{\mu}} -\frac{\partial F}{\partial S} p_{\mu}\tag{8}$$

$$\frac{\partial x_{\mu}}{\partial \tau}= \frac{\partial F}{\partial p^{\mu}}. \tag{9}$$

Remark: HJ theory says $$p_{\mu}= \frac{\partial S}{\partial x^{\mu}}.\tag{10}$$

The problem is to derive from here the equation for Lorentz force

  • 1
    $\begingroup$ Suggestion: Replace the word Lorentz equation with the word Lorentz force. $\endgroup$
    – Qmechanic
    Oct 29, 2018 at 13:15
  • $\begingroup$ Comments to the post (v6): 1. Eq. (6) seems wrong for dimensional reasons. 2. Double-check indices in eq.(7). $\endgroup$
    – Qmechanic
    Oct 30, 2018 at 14:46

1 Answer 1


This answer does not address OP's specific question about the method of characteristics, but sketches a systematic derivation (of the various equations involved) starting from a Lagrangian formulation.

  1. A Lagrangian for a relativistic point particle of mass $m$ and charge $q$ in a EM background $A_{\mu}$ and gravitational background $g_{\mu\nu}$ is$^1$ $$ L~:=~L_0 - U,\qquad L_0~:=~\pm \frac{\dot{x}^2}{2e}-\frac{e m^2}{2}, \qquad \dot{x}^2~:=~g_{\mu\nu}(x)~ \dot{x}^{\mu}\dot{x}^{\nu}, \qquad \dot {x}^{\mu} ~:=~\frac{dx^{\mu}}{d\tau},\tag{A}$$ with Minkowski sign convention $(\mp,\pm,\pm,\pm)$ and speed-of-light $c=1$. Here $\tau$ is the world-line (WL) parameter (which is not necessarily proper time) and $e>0$ is an einbein field. The velocity-dependent Lorentz potential is $$ U~:=~ \mp q{\dot x}^{\mu} A_{\mu}, \tag{B} $$ with corresponding generalized Lorentz 4-force$^2$ $$ F_{\mu}~:=~\frac{d}{d\tau} \frac{\partial U}{\partial \dot{x}^{\mu}} - \frac{\partial U}{\partial x^{\mu}}~=~\pm qF_{\mu\nu}\dot {x}^{\nu}, \qquad F_{\mu\nu}~:=~\partial_{\mu}A_{\nu}-\partial_{\nu}A_{\mu}. \tag{C}$$

  2. The canonical/conjugate 4-momentum$^2$ becomes $$ p_{\mu}~:=~\frac{\partial L}{\partial \dot{x}^{\mu}}~=~\pm\frac{g_{\mu\nu}\dot{x}^{\nu}}{e}\pm q A_{\mu}.\tag{D}$$

  3. After a Legendre transformation the Hamiltonian reads $$ H~=~\frac{e}{2}\left(m^2\pm ( p\mp qA)^2 \right). \tag{E}$$ Note that the $e$ field is a Lagrange multiplier for the mass-shell constraint.

  4. The Hamilton-Jacobi (HJ) equation becomes essentially the mass-shell constraint $$ 0~=~E~=~\frac{e}{2}\left(m^2\pm ( \frac{\partial W}{\partial x}-qA)^2 \right), \tag{F}$$ where $W$ is Hamilton's characteristic function, and$^3$ $$ p_{\mu}~=~\pm \frac{\partial W}{\partial x^{\mu}}. \tag{G}$$ The fact that the energy $E$ is zero can be viewed as a consequence of WL reparametrization invariance $\tau\to \tau^{\prime}=f(\tau)$.


$^1$ To achieve the standard square root Lagrangian, simply integrate out the $e$ field, cf. e.g. this Phys.SE post.

$^2$ The usual notions of 4-momentum & 4-force correspond to the gauge where the world-line (WL) parameter $\tau$ is proper time.

$^3$ For sign conventions, see also this Phys.SE post.


Your Answer

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

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