Hamilton–Jacobi equation

The Hamilton–Jacobi equation is a first-order, non-linear partial differential equation

for a system of particles at coordinates ⁠${\displaystyle \mathbf {q} }$ ⁠. The function ${\displaystyle H}$  is the system's Hamiltonian giving the system's energy. The solution of the equation is the action functional, ⁠${\displaystyle S}$ ⁠,^[4] called Hamilton's principal function in older textbooks. The solution can be related to the system Lagrangian ${\displaystyle \ {\mathcal {L}}\ }$  by an indefinite integral of the form used in the principle of least action:^[5]: 431 $${\displaystyle \ S=\int {\mathcal {L}}\ \operatorname {d} t+~{\mathsf {some\ constant}}~}$$  Geometrical surfaces of constant action are perpendicular to system trajectories, creating a wavefront-like view of the system dynamics. This property of the Hamilton–Jacobi equation connects classical mechanics to quantum mechanics.^[6]: 175

Boldface variables such as ${\displaystyle \mathbf {q} }$  represent a list of ${\displaystyle N}$  generalized coordinates, $${\displaystyle \mathbf {q} =(q_{1},q_{2},\ldots ,q_{N-1},q_{N})}$$ 

A dot over a variable or list signifies the time derivative (see Newton's notation). For example, $${\displaystyle {\dot {\mathbf {q} }}={\frac {d\mathbf {q} }{dt}}.}$$ 

The dot product notation between two lists of the same number of coordinates is a shorthand for the sum of the products of corresponding components, such as $${\displaystyle \mathbf {p} \cdot \mathbf {q} =\sum _{k=1}^{N}p_{k}q_{k}.}$$ 

Let the Hessian matrix ${\textstyle H_{\cal {L}}(\mathbf {q} ,\mathbf {\dot {q}} ,t)=\left\{\partial ^{2}{\cal {L}}/\partial {\dot {q}}^{i}\partial {\dot {q}}^{j}\right\}_{ij}}$  be invertible. The relation $${\displaystyle {\frac {d}{dt}}{\frac {\partial {\cal {L}}}{\partial {\dot {q}}^{i}}}=\sum _{j=1}^{n}\left({\frac {\partial ^{2}{\cal {L}}}{\partial {\dot {q}}^{i}\partial {\dot {q}}^{j}}}{\ddot {q}}^{j}+{\frac {\partial ^{2}{\cal {L}}}{\partial {\dot {q}}^{i}\partial {q}^{j}}}{\dot {q}}^{j}\right)+{\frac {\partial ^{2}{\cal {L}}}{\partial {\dot {q}}^{i}\partial t}},\qquad i=1,\ldots ,n,}$$  shows that the Euler–Lagrange equations form a ${\displaystyle n\times n}$  system of second-order ordinary differential equations. Inverting the matrix ${\displaystyle H_{\cal {L}}}$  transforms this system into $${\displaystyle {\ddot {q}}^{i}=F_{i}(\mathbf {q} ,\mathbf {\dot {q}} ,t),\ i=1,\ldots ,n.}$$ 

Let a time instant ${\displaystyle t_{0}}$  and a point ${\displaystyle \mathbf {q} _{0}\in M}$  in the configuration space be fixed. The existence and uniqueness theorems guarantee that, for every ${\displaystyle \mathbf {v} _{0},}$  the initial value problem with the conditions ${\displaystyle \gamma |_{\tau =t_{0}}=\mathbf {q} _{0}}$  and ${\displaystyle {\dot {\gamma }}|_{\tau =t_{0}}=\mathbf {v} _{0}}$  has a locally unique solution ${\displaystyle \gamma =\gamma (\tau ;t_{0},\mathbf {q} _{0},\mathbf {v} _{0}).}$  Additionally, let there be a sufficiently small time interval ${\displaystyle (t_{0},t_{1})}$  such that extremals with different initial velocities ${\displaystyle \mathbf {v} _{0}}$  would not intersect in ${\displaystyle M\times (t_{0},t_{1}).}$  The latter means that, for any ${\displaystyle \mathbf {q} \in M}$  and any ${\displaystyle t\in (t_{0},t_{1}),}$  there can be at most one extremal ${\displaystyle \gamma =\gamma (\tau ;t,t_{0},\mathbf {q} ,\mathbf {q} _{0})}$  for which ${\displaystyle \gamma |_{\tau =t_{0}}=\mathbf {q} _{0}}$  and ${\displaystyle \gamma |_{\tau =t}=\mathbf {q} .}$  Substituting ${\displaystyle \gamma =\gamma (\tau ;t,t_{0},\mathbf {q} ,\mathbf {q} _{0})}$  into the action functional results in the Hamilton's principal function (HPF)

where

• ${\displaystyle \gamma =\gamma (\tau ;t,t_{0},\mathbf {q} ,\mathbf {q} _{0}),}$ 
• ${\displaystyle \gamma |_{\tau =t_{0}}=\mathbf {q} _{0},}$ 
• ${\displaystyle \gamma |_{\tau =t}=\mathbf {q} .}$ 

The momenta are defined as the quantities ${\textstyle p_{i}(\mathbf {q} ,\mathbf {\dot {q}} ,t)=\partial {\cal {L}}/\partial {\dot {q}}^{i}.}$  This section shows that the dependency of ${\displaystyle p_{i}}$  on ${\displaystyle \mathbf {\dot {q}} }$  disappears, once the HPF is known.

Indeed, let a time instant ${\displaystyle t_{0}}$  and a point ${\displaystyle \mathbf {q} _{0}}$  in the configuration space be fixed. For every time instant ${\displaystyle t}$  and a point ${\displaystyle \mathbf {q} ,}$  let ${\displaystyle \gamma =\gamma (\tau ;t,t_{0},\mathbf {q} ,\mathbf {q} _{0})}$  be the (unique) extremal from the definition of the Hamilton's principal function ⁠${\displaystyle S}$ ⁠. Call ${\displaystyle \mathbf {v} \,{\stackrel {\text{def}}{=}}\,{\dot {\gamma }}(\tau ;t,t_{0},\mathbf {q} ,\mathbf {q} _{0})|_{\tau =t}}$  the velocity at ⁠${\displaystyle \tau =t}$ ⁠. Then

Given the Hamiltonian ${\displaystyle H(\mathbf {q} ,\mathbf {p} ,t)}$  of a mechanical system, the Hamilton–Jacobi equation is a first-order, non-linear partial differential equation for the Hamilton's principal function ${\displaystyle S}$ ,^[7]

Alternatively, as described below, the Hamilton–Jacobi equation may be derived from Hamiltonian mechanics by treating ${\displaystyle S}$  as the generating function for a canonical transformation of the classical Hamiltonian $${\displaystyle H=H(q_{1},q_{2},\ldots ,q_{N};p_{1},p_{2},\ldots ,p_{N};t).}$$ 

The conjugate momenta correspond to the first derivatives of ${\displaystyle S}$  with respect to the generalized coordinates $${\displaystyle p_{k}={\frac {\partial S}{\partial q_{k}}}.}$$ 

As a solution to the Hamilton–Jacobi equation, the principal function contains ${\displaystyle N+1}$  undetermined constants, the first ${\displaystyle N}$  of them denoted as ${\displaystyle \alpha _{1},\,\alpha _{2},\dots ,\alpha _{N}}$ , and the last one coming from the integration of ${\displaystyle {\frac {\partial S}{\partial t}}}$ .

The relationship between ${\displaystyle \mathbf {p} }$  and ${\displaystyle \mathbf {q} }$  then describes the orbit in phase space in terms of these constants of motion. Furthermore, the quantities $${\displaystyle \beta _{k}={\frac {\partial S}{\partial \alpha _{k}}},\quad k=1,2,\ldots ,N}$$  are also constants of motion, and these equations can be inverted to find ${\displaystyle \mathbf {q} }$  as a function of all the ${\displaystyle \alpha }$  and ${\displaystyle \beta }$  constants and time.^[8]

The Hamilton–Jacobi equation is a single, first-order partial differential equation for the function of the ${\displaystyle N}$  generalized coordinates ${\displaystyle q_{1},\,q_{2},\dots ,q_{N}}$  and the time ${\displaystyle t}$ . The generalized momenta do not appear, except as derivatives of ${\displaystyle S}$ , the classical action.

For comparison, in the equivalent Euler–Lagrange equations of motion of Lagrangian mechanics, the conjugate momenta also do not appear; however, those equations are a system of ${\displaystyle N}$ , generally second-order equations for the time evolution of the generalized coordinates. Similarly, Hamilton's equations of motion are another system of 2N first-order equations for the time evolution of the generalized coordinates and their conjugate momenta ${\displaystyle p_{1},\,p_{2},\dots ,p_{N}}$ .

Since the HJE is an equivalent expression of an integral minimization problem such as Hamilton's principle, the HJE can be useful in other problems of the calculus of variations and, more generally, in other branches of mathematics and physics, such as dynamical systems, symplectic geometry and quantum chaos. For example, the Hamilton–Jacobi equations can be used to determine the geodesics on a Riemannian manifold, an important variational problem in Riemannian geometry. However as a computational tool, the partial differential equations are notoriously complicated to solve except when is it possible to separate the independent variables; in this case the HJE become computationally useful.^[5]: 444

Any canonical transformation involving a type-2 generating function ${\displaystyle G_{2}(\mathbf {q} ,\mathbf {P} ,t)}$  leads to the relations $${\displaystyle \mathbf {p} ={\partial G_{2} \over \partial \mathbf {q} },\quad \mathbf {Q} ={\partial G_{2} \over \partial \mathbf {P} },\quad K(\mathbf {Q} ,\mathbf {P} ,t)=H(\mathbf {q} ,\mathbf {p} ,t)+{\partial G_{2} \over \partial t}}$$  and Hamilton's equations in terms of the new variables ${\displaystyle \mathbf {P} ,\,\mathbf {Q} }$  and new Hamiltonian ${\displaystyle K}$  have the same form: $${\displaystyle {\dot {\mathbf {P} }}=-{\partial K \over \partial \mathbf {Q} },\quad {\dot {\mathbf {Q} }}=+{\partial K \over \partial \mathbf {P} }.}$$ 

To derive the HJE, a generating function ${\displaystyle G_{2}(\mathbf {q} ,\mathbf {P} ,t)}$  is chosen in such a way that, it will make the new Hamiltonian ${\displaystyle K=0}$ . Hence, all its derivatives are also zero, and the transformed Hamilton's equations become trivial $${\displaystyle {\dot {\mathbf {P} }}={\dot {\mathbf {Q} }}=0}$$  so the new generalized coordinates and momenta are constants of motion. As they are constants, in this context the new generalized momenta ${\displaystyle \mathbf {P} }$  are usually denoted ${\displaystyle \alpha _{1},\,\alpha _{2},\dots ,\alpha _{N}}$ , i.e. ${\displaystyle P_{m}=\alpha _{m}}$  and the new generalized coordinates ${\displaystyle \mathbf {Q} }$  are typically denoted as ${\displaystyle \beta _{1},\,\beta _{2},\dots ,\beta _{N}}$ , so ${\displaystyle Q_{m}=\beta _{m}}$ .

Setting the generating function equal to Hamilton's principal function, plus an arbitrary constant ${\displaystyle A}$ : $${\displaystyle G_{2}(\mathbf {q} ,{\boldsymbol {\alpha }},t)=S(\mathbf {q} ,t)+A,}$$  the HJE automatically arises $${\displaystyle \mathbf {p} ={\frac {\partial G_{2}}{\partial \mathbf {q} }}={\frac {\partial S}{\partial \mathbf {q} }}\,\rightarrow \,H(\mathbf {q} ,\mathbf {p} ,t)+{\partial G_{2} \over \partial t}=0\,\rightarrow \,H\left(\mathbf {q} ,{\frac {\partial S}{\partial \mathbf {q} }},t\right)+{\partial S \over \partial t}=0.}$$ 

When solved for ${\displaystyle S(\mathbf {q} ,{\boldsymbol {\alpha }},t)}$ , these also give us the useful equations $${\displaystyle \mathbf {Q} ={\boldsymbol {\beta }}={\partial S \over \partial {\boldsymbol {\alpha }}},}$$  or written in components for clarity $${\displaystyle Q_{m}=\beta _{m}={\frac {\partial S(\mathbf {q} ,{\boldsymbol {\alpha }},t)}{\partial \alpha _{m}}}.}$$ 

Ideally, these N equations can be inverted to find the original generalized coordinates ${\displaystyle \mathbf {q} }$  as a function of the constants ${\displaystyle {\boldsymbol {\alpha }},\,{\boldsymbol {\beta }},}$  and ${\displaystyle t}$ , thus solving the original problem.

When the problem allows additive separation of variables, the HJE leads directly to constants of motion. For example, the time t can be separated if the Hamiltonian does not depend on time explicitly. In that case, the time derivative ${\displaystyle {\frac {\partial S}{\partial t}}}$  in the HJE must be a constant, usually denoted (${\displaystyle -E}$ ), giving the separated solution $${\displaystyle S=W(q_{1},q_{2},\ldots ,q_{N})-Et}$$  where the time-independent function ${\displaystyle W(\mathbf {q} )}$  is sometimes called the abbreviated action or Hamilton's characteristic function ^[5]: 434 and sometimes^[9]: 607 written ${\displaystyle S_{0}}$  (see action principle names). The reduced Hamilton–Jacobi equation can then be written $${\displaystyle H\left(\mathbf {q} ,{\frac {\partial S}{\partial \mathbf {q} }}\right)=E.}$$ 

To illustrate separability for other variables, a certain generalized coordinate ${\displaystyle q_{k}}$  and its derivative ${\displaystyle {\frac {\partial S}{\partial q_{k}}}}$  are assumed to appear together as a single function $${\displaystyle \psi \left(q_{k},{\frac {\partial S}{\partial q_{k}}}\right)}$$  in the Hamiltonian $${\displaystyle H=H(q_{1},q_{2},\ldots ,q_{k-1},q_{k+1},\ldots ,q_{N};p_{1},p_{2},\ldots ,p_{k-1},p_{k+1},\ldots ,p_{N};\psi ;t).}$$ 

In that case, the function S can be partitioned into two functions, one that depends only on q_k and another that depends only on the remaining generalized coordinates $${\displaystyle S=S_{k}(q_{k})+S_{\text{rem}}(q_{1},\ldots ,q_{k-1},q_{k+1},\ldots ,q_{N},t).}$$ 

Substitution of these formulae into the Hamilton–Jacobi equation shows that the function ψ must be a constant (denoted here as ${\displaystyle \Gamma _{k}}$ ), yielding a first-order ordinary differential equation for ${\displaystyle S_{k}(q_{k}),}$ 

$${\displaystyle \psi \left(q_{k},{\frac {dS_{k}}{dq_{k}}}\right)=\Gamma _{k}.}$$ 

In fortunate cases, the function ${\displaystyle S}$  can be separated completely into ${\displaystyle N}$  functions ${\displaystyle S_{m}(q_{m}),}$  $${\displaystyle S=S_{1}(q_{1})+S_{2}(q_{2})+\cdots +S_{N}(q_{N})-Et.}$$ 

In such a case, the problem devolves to ${\displaystyle N}$  ordinary differential equations.

The separability of S depends both on the Hamiltonian and on the choice of generalized coordinates. For orthogonal coordinates and Hamiltonians that have no time dependence and are quadratic in the generalized momenta, ${\displaystyle S}$  will be completely separable if the potential energy is additively separable in each coordinate, where the potential energy term for each coordinate is multiplied by the coordinate-dependent factor in the corresponding momentum term of the Hamiltonian (the Staeckel conditions). For illustration, several examples in orthogonal coordinates are worked in the next sections.

In spherical coordinates the Hamiltonian of a free particle moving in a conservative potential U can be written $${\displaystyle H={\frac {1}{2m}}\left[p_{r}^{2}+{\frac {p_{\theta }^{2}}{r^{2}}}+{\frac {p_{\phi }^{2}}{r^{2}\sin ^{2}\theta }}\right]+U(r,\theta ,\phi ).}$$ 

The Hamilton–Jacobi equation is completely separable in these coordinates provided that there exist functions ${\displaystyle U_{r}(r),U_{\theta }(\theta ),U_{\phi }(\phi )}$  such that ${\displaystyle U}$  can be written in the analogous form $${\displaystyle U(r,\theta ,\phi )=U_{r}(r)+{\frac {U_{\theta }(\theta )}{r^{2}}}+{\frac {U_{\phi }(\phi )}{r^{2}\sin ^{2}\theta }}.}$$ 

Substitution of the completely separated solution $${\displaystyle S=S_{r}(r)+S_{\theta }(\theta )+S_{\phi }(\phi )-Et}$$  into the HJE yields $${\displaystyle {\frac {1}{2m}}\left({\frac {dS_{r}}{dr}}\right)^{2}+U_{r}(r)+{\frac {1}{2mr^{2}}}\left[\left({\frac {dS_{\theta }}{d\theta }}\right)^{2}+2mU_{\theta }(\theta )\right]+{\frac {1}{2mr^{2}\sin ^{2}\theta }}\left[\left({\frac {dS_{\phi }}{d\phi }}\right)^{2}+2mU_{\phi }(\phi )\right]=E.}$$ 

This equation may be solved by successive integrations of ordinary differential equations, beginning with the equation for ${\displaystyle \phi }$  $${\displaystyle \left({\frac {dS_{\phi }}{d\phi }}\right)^{2}+2mU_{\phi }(\phi )=\Gamma _{\phi }}$$  where ${\displaystyle \Gamma _{\phi }}$  is a constant of the motion that eliminates the ${\displaystyle \phi }$  dependence from the Hamilton–Jacobi equation $${\displaystyle {\frac {1}{2m}}\left({\frac {dS_{r}}{dr}}\right)^{2}+U_{r}(r)+{\frac {1}{2mr^{2}}}\left[{\frac {1}{\sin ^{2}\theta }}\left({\frac {dS_{\theta }}{d\theta }}\right)^{2}+{\frac {2m}{\sin ^{2}\theta }}U_{\theta }(\theta )+\Gamma _{\phi }\right]=E.}$$ 

The next ordinary differential equation involves the ${\displaystyle \theta }$  generalized coordinate $${\displaystyle {\frac {1}{\sin ^{2}\theta }}\left({\frac {dS_{\theta }}{d\theta }}\right)^{2}+{\frac {2m}{\sin ^{2}\theta }}U_{\theta }(\theta )+\Gamma _{\phi }=\Gamma _{\theta }}$$  where ${\displaystyle \Gamma _{\theta }}$  is again a constant of the motion that eliminates the ${\displaystyle \theta }$  dependence and reduces the HJE to the final ordinary differential equation $${\displaystyle {\frac {1}{2m}}\left({\frac {dS_{r}}{dr}}\right)^{2}+U_{r}(r)+{\frac {\Gamma _{\theta }}{2mr^{2}}}=E}$$  whose integration completes the solution for ${\displaystyle S}$ .

The Hamiltonian in elliptic cylindrical coordinates can be written $${\displaystyle H={\frac {p_{\mu }^{2}+p_{\nu }^{2}}{2ma^{2}\left(\sinh ^{2}\mu +\sin ^{2}\nu \right)}}+{\frac {p_{z}^{2}}{2m}}+U(\mu ,\nu ,z)}$$  where the foci of the ellipses are located at ${\displaystyle \pm a}$  on the ${\displaystyle x}$ -axis. The Hamilton–Jacobi equation is completely separable in these coordinates provided that ${\displaystyle U}$  has an analogous form $${\displaystyle U(\mu ,\nu ,z)={\frac {U_{\mu }(\mu )+U_{\nu }(\nu )}{\sinh ^{2}\mu +\sin ^{2}\nu }}+U_{z}(z)}$$  where ${\displaystyle U_{\mu }(\mu )}$ , ${\displaystyle U_{\nu }(\nu )}$  and ${\displaystyle U_{z}(z)}$  are arbitrary functions. Substitution of the completely separated solution $${\displaystyle S=S_{\mu }(\mu )+S_{\nu }(\nu )+S_{z}(z)-Et}$$  into the HJE yields $${\displaystyle {\frac {1}{2m}}\left({\frac {dS_{z}}{dz}}\right)^{2}+U_{z}(z)+{\frac {1}{2ma^{2}\left(\sinh ^{2}\mu +\sin ^{2}\nu \right)}}\left[\left({\frac {dS_{\mu }}{d\mu }}\right)^{2}+\left({\frac {dS_{\nu }}{d\nu }}\right)^{2}+2ma^{2}U_{\mu }(\mu )+2ma^{2}U_{\nu }(\nu )\right]=E.}$$ 

Separating the first ordinary differential equation $${\displaystyle {\frac {1}{2m}}\left({\frac {dS_{z}}{dz}}\right)^{2}+U_{z}(z)=\Gamma _{z}}$$  yields the reduced Hamilton–Jacobi equation (after re-arrangement and multiplication of both sides by the denominator) $${\displaystyle \left({\frac {dS_{\mu }}{d\mu }}\right)^{2}+\left({\frac {dS_{\nu }}{d\nu }}\right)^{2}+2ma^{2}U_{\mu }(\mu )+2ma^{2}U_{\nu }(\nu )=2ma^{2}\left(\sinh ^{2}\mu +\sin ^{2}\nu \right)\left(E-\Gamma _{z}\right)}$$  which itself may be separated into two independent ordinary differential equations $${\displaystyle \left({\frac {dS_{\mu }}{d\mu }}\right)^{2}+2ma^{2}U_{\mu }(\mu )+2ma^{2}\left(\Gamma _{z}-E\right)\sinh ^{2}\mu =\Gamma _{\mu }}$$  $${\displaystyle \left({\frac {dS_{\nu }}{d\nu }}\right)^{2}+2ma^{2}U_{\nu }(\nu )+2ma^{2}\left(\Gamma _{z}-E\right)\sin ^{2}\nu =\Gamma _{\nu }}$$  that, when solved, provide a complete solution for ${\displaystyle S}$ .

The Hamiltonian in parabolic cylindrical coordinates can be written $${\displaystyle H={\frac {p_{\sigma }^{2}+p_{\tau }^{2}}{2m\left(\sigma ^{2}+\tau ^{2}\right)}}+{\frac {p_{z}^{2}}{2m}}+U(\sigma ,\tau ,z).}$$ 

The Hamilton–Jacobi equation is completely separable in these coordinates provided that ${\displaystyle U}$  has an analogous form $${\displaystyle U(\sigma ,\tau ,z)={\frac {U_{\sigma }(\sigma )+U_{\tau }(\tau )}{\sigma ^{2}+\tau ^{2}}}+U_{z}(z)}$$  where ${\displaystyle U_{\sigma }(\sigma )}$ , ${\displaystyle U_{\tau }(\tau )}$ , and ${\displaystyle U_{z}(z)}$  are arbitrary functions. Substitution of the completely separated solution $${\displaystyle S=S_{\sigma }(\sigma )+S_{\tau }(\tau )+S_{z}(z)-Et+{\text{constant}}}$$  into the HJE yields $${\displaystyle {\frac {1}{2m}}\left({\frac {dS_{z}}{dz}}\right)^{2}+U_{z}(z)+{\frac {1}{2m\left(\sigma ^{2}+\tau ^{2}\right)}}\left[\left({\frac {dS_{\sigma }}{d\sigma }}\right)^{2}+\left({\frac {dS_{\tau }}{d\tau }}\right)^{2}+2mU_{\sigma }(\sigma )+2mU_{\tau }(\tau )\right]=E.}$$ 

Separating the first ordinary differential equation $${\displaystyle {\frac {1}{2m}}\left({\frac {dS_{z}}{dz}}\right)^{2}+U_{z}(z)=\Gamma _{z}}$$  yields the reduced Hamilton–Jacobi equation (after re-arrangement and multiplication of both sides by the denominator) $${\displaystyle \left({\frac {dS_{\sigma }}{d\sigma }}\right)^{2}+\left({\frac {dS_{\tau }}{d\tau }}\right)^{2}+2mU_{\sigma }(\sigma )+2mU_{\tau }(\tau )=2m\left(\sigma ^{2}+\tau ^{2}\right)\left(E-\Gamma _{z}\right)}$$  which itself may be separated into two independent ordinary differential equations $${\displaystyle \left({\frac {dS_{\sigma }}{d\sigma }}\right)^{2}+2mU_{\sigma }(\sigma )+2m\sigma ^{2}\left(\Gamma _{z}-E\right)=\Gamma _{\sigma }}$$  $${\displaystyle \left({\frac {dS_{\tau }}{d\tau }}\right)^{2}+2mU_{\tau }(\tau )+2m\tau ^{2}\left(\Gamma _{z}-E\right)=\Gamma _{\tau }}$$  that, when solved, provide a complete solution for ${\displaystyle S}$ .

The HJE establishes a duality between trajectories and wavefronts.^[10] For example, in geometrical optics, light can be considered either as “rays” or waves. The wave front can be defined as the surface ${\textstyle {\cal {C}}_{t}}$  that the light emitted at time ${\textstyle t=0}$  has reached at time ${\textstyle t}$ . Light rays and wave fronts are dual: if one is known, the other can be deduced.

More precisely, geometrical optics is a variational problem where the “action” is the travel time ${\textstyle T}$  along a path,$${\displaystyle T={\frac {1}{c}}\int _{A}^{B}n\,ds}$$  where ${\textstyle n}$  is the medium's index of refraction and ${\textstyle ds}$  is an infinitesimal arc length. From the above formulation, one can compute the ray paths using the Euler–Lagrange formulation; alternatively, one can compute the wave fronts by solving the Hamilton–Jacobi equation. Knowing one leads to knowing the other.

The above duality is very general and applies to all systems that derive from a variational principle: either compute the trajectories using Euler–Lagrange equations or the wave fronts by using Hamilton–Jacobi equation.

The wave front at time ${\textstyle t}$ , for a system initially at ${\textstyle \mathbf {q} _{0}}$  at time ${\textstyle t_{0}}$ , is defined as the collection of points ${\textstyle \mathbf {q} }$  such that ${\textstyle S(\mathbf {q} ,t)={\text{const}}}$ . If ${\textstyle S(\mathbf {q} ,t)}$  is known, the momentum is immediately deduced.$${\displaystyle \mathbf {p} ={\frac {\partial S}{\partial \mathbf {q} }}.}$$ 

Once ${\textstyle \mathbf {p} }$  is known, tangents to the trajectories ${\textstyle {\dot {\mathbf {q} }}}$  are computed by solving the equation$${\displaystyle {\frac {\partial {\cal {L}}}{\partial {\dot {\mathbf {q} }}}}={\boldsymbol {p}}}$$ for ${\textstyle {\dot {\mathbf {q} }}}$ , where ${\textstyle {\cal {L}}}$  is the Lagrangian. The trajectories are then recovered from the knowledge of ${\textstyle {\dot {\mathbf {q} }}}$ .

The isosurfaces of the function ${\displaystyle S(\mathbf {q} ,t)}$  can be determined at any time t. The motion of an ${\displaystyle S}$ -isosurface as a function of time is defined by the motions of the particles beginning at the points ${\displaystyle \mathbf {q} }$  on the isosurface. The motion of such an isosurface can be thought of as a wave moving through ${\displaystyle \mathbf {q} }$ -space, although it does not obey the wave equation exactly. To show this, let S represent the phase of a wave $${\displaystyle \psi =\psi _{0}e^{iS/\hbar }}$$  where ${\displaystyle \hbar }$  is a constant (the Planck constant) introduced to make the exponential argument dimensionless; changes in the amplitude of the wave can be represented by having ${\displaystyle S}$  be a complex number. The Hamilton–Jacobi equation is then rewritten as $${\displaystyle {\frac {\hbar ^{2}}{2m}}\nabla ^{2}\psi -U\psi ={\frac {\hbar }{i}}{\frac {\partial \psi }{\partial t}}}$$  which is the Schrödinger equation.

Conversely, starting with the Schrödinger equation and our ansatz for ${\displaystyle \psi }$ , it can be deduced that^[11] $${\displaystyle {\frac {1}{2m}}\left(\nabla S\right)^{2}+U+{\frac {\partial S}{\partial t}}={\frac {i\hbar }{2m}}\nabla ^{2}S.}$$ 

The classical limit (${\displaystyle \hbar \rightarrow 0}$ ) of the Schrödinger equation above becomes identical to the following variant of the Hamilton–Jacobi equation, $${\displaystyle {\frac {1}{2m}}\left(\nabla S\right)^{2}+U+{\frac {\partial S}{\partial t}}=0.}$$ 

Using the energy–momentum relation in the form^[12] $${\displaystyle g^{\alpha \beta }P_{\alpha }P_{\beta }-(mc)^{2}=0}$$  for a particle of rest mass ${\displaystyle m}$  travelling in curved space, where ${\displaystyle g^{\alpha \beta }}$  are the contravariant coordinates of the metric tensor (i.e., the inverse metric) solved from the Einstein field equations, and ${\displaystyle c}$  is the speed of light. Setting the four-momentum ${\displaystyle P_{\alpha }}$  equal to the four-gradient of the action ${\displaystyle S}$ , $${\displaystyle P_{\alpha }=-{\frac {\partial S}{\partial x^{\alpha }}}}$$  gives the Hamilton–Jacobi equation in the geometry determined by the metric ${\displaystyle g}$ : $${\displaystyle g^{\alpha \beta }{\frac {\partial S}{\partial x^{\alpha }}}{\frac {\partial S}{\partial x^{\beta }}}-(mc)^{2}=0,}$$  in other words, in a gravitational field.

For a particle of rest mass ${\displaystyle m}$  and electric charge ${\displaystyle e}$  moving in electromagnetic field with four-potential ${\displaystyle A_{i}=(\phi ,\mathrm {A} )}$  in vacuum, the Hamilton–Jacobi equation in geometry determined by the metric tensor ${\displaystyle g^{ik}=g_{ik}}$  has a form $${\displaystyle g^{ik}\left({\frac {\partial S}{\partial x^{i}}}+{\frac {e}{c}}A_{i}\right)\left({\frac {\partial S}{\partial x^{k}}}+{\frac {e}{c}}A_{k}\right)=m^{2}c^{2}}$$  and can be solved for the Hamilton principal action function ${\displaystyle S}$  to obtain further solution for the particle trajectory and momentum:^[13] $${\displaystyle x=-{\frac {e}{c\gamma }}\int A_{z}\,d\xi ,}$$  $${\displaystyle y=-{\frac {e}{c\gamma }}\int A_{y}\,d\xi ,}$$  $${\displaystyle z=-{\frac {e^{2}}{2c^{2}\gamma ^{2}}}\int (\mathrm {A} ^{2}-{\overline {\mathrm {A} ^{2}}})\,d\xi ,}$$  $${\displaystyle \xi =ct-{\frac {e^{2}}{2\gamma ^{2}c^{2}}}\int (\mathrm {A} ^{2}-{\overline {\mathrm {A} ^{2}}})\,d\xi ,}$$  $${\displaystyle p_{x}=-{\frac {e}{c}}A_{x},\quad p_{y}=-{\frac {e}{c}}A_{y},}$$  $${\displaystyle p_{z}={\frac {e^{2}}{2\gamma c}}(\mathrm {A} ^{2}-{\overline {\mathrm {A} ^{2}}}),}$$  $${\displaystyle {\mathcal {E}}=c\gamma +{\frac {e^{2}}{2\gamma c}}(\mathrm {A} ^{2}-{\overline {\mathrm {A} ^{2}}}),}$$  where ${\displaystyle \xi =ct-z}$  and ${\displaystyle \gamma ^{2}=m^{2}c^{2}+{\frac {e^{2}}{c^{2}}}{\overline {A}}^{2}}$  with ${\displaystyle {\overline {\mathbf {A} }}}$  the cycle average of the vector potential.

In the case of circular polarization, $${\displaystyle E_{x}=E_{0}\sin \omega \xi _{1},\quad E_{y}=E_{0}\cos \omega \xi _{1},}$$  $${\displaystyle A_{x}={\frac {cE_{0}}{\omega }}\cos \omega \xi _{1},\quad A_{y}=-{\frac {cE_{0}}{\omega }}\sin \omega \xi _{1}.}$$ 

Hence $${\displaystyle x=-{\frac {ecE_{0}}{\omega }}\sin \omega \xi _{1},}$$  $${\displaystyle y=-{\frac {ecE_{0}}{\omega }}\cos \omega \xi _{1},}$$  $${\displaystyle p_{x}=-{\frac {eE_{0}}{\omega }}\cos \omega \xi _{1},}$$  $${\displaystyle p_{y}={\frac {eE_{0}}{\omega }}\sin \omega \xi _{1},}$$  where ${\displaystyle \xi _{1}=\xi /c}$ , implying the particle moving along a circular trajectory with a permanent radius ${\displaystyle ecE_{0}/\gamma \omega ^{2}}$  and an invariable value of momentum ${\displaystyle eE_{0}/\omega ^{2}}$  directed along a magnetic field vector.

For the flat, monochromatic, linearly polarized wave with a field ${\displaystyle E}$  directed along the axis ${\displaystyle y}$  $${\displaystyle E_{y}=E_{0}\cos \omega \xi _{1},}$$  $${\displaystyle A_{y}=-{\frac {cE_{0}}{\omega }}\sin \omega \xi _{1},}$$  hence $${\displaystyle x={\text{const}},}$$  $${\displaystyle y_{0}=-{\frac {ecE_{0}}{\gamma \omega ^{2}}},}$$  $${\displaystyle y=y_{0}\cos \omega \xi _{1},\quad z=C_{z}y_{0}\sin 2\omega \xi _{1},}$$  $${\displaystyle C_{z}={\frac {eE_{0}}{8\gamma \omega }},\quad \gamma ^{2}=m^{2}c^{2}+{\frac {e^{2}E_{0}^{2}}{2\omega ^{2}}},}$$  $${\displaystyle p_{x}=0,}$$  $${\displaystyle p_{y,0}={\frac {eE_{0}}{\omega }},}$$  $${\displaystyle p_{y}=p_{y,0}\sin \omega \xi _{1},}$$  $${\displaystyle p_{z}=-2C_{z}p_{y,0}\cos 2\omega \xi _{1}}$$  implying the particle figure-8 trajectory with a long its axis oriented along the electric field ${\displaystyle E}$  vector.