A note on the equivalence of Lagrangian and Hamiltonian formulations at post-Newtonian approximations

Abstract

It was claimed recently that a low order post-Newtonian (PN) Lagrangian formulation, which corresponds to the Euler-Lagrange equations up to an infinite PN order, can be identical to a PN Hamiltonian formulation at the infinite order from a theoretical point of view. This result is difficult to check because in most cases one does not know what both the Euler-Lagrange equations and the equivalent Hamiltonian are at the infinite order. However, no difficulty exists for a special 1PN Lagrangian formulation of relativistic circular restricted three-body problem, where both the Euler-Lagrange equations and the equivalent Hamiltonian not only are expanded to all PN orders but also have converged functions. Consequently, the analytical evidence supports this claim. As far as numerical evidences are concerned, the Hamiltonian equivalent to the Euler-Lagrange equations for the lower order Lagrangian requires that they both be only at higher enough finite orders.