Modified Semi-Classical Methods for Nonlinear Quantum Oscillations Problems

TL;DR

This paper introduces a modified semi-classical method for solving nonlinear quantum oscillation problems, providing new analytical tools and accurate eigenfunction approximations, especially for anharmonic oscillators.

Contribution

It develops a novel semi-classical approach replacing the Hamilton-Jacobi equation with an inverted potential variant, and proves the existence of a fundamental solution under certain conditions.

Findings

Eigenvalue expansions match Rayleigh-Schrodinger theory for anharmonic oscillators

Wave functions decay more rapidly than Gaussian, improving accuracy

Ground state energy expansions are likely Borel summable

Abstract

We develop a modified semi-classical approach to the approximate solution of Schrodinger's equation for certain nonlinear quantum oscillations problems. At lowest order, the Hamilton-Jacobi equation of the conventional semi-classical formalism is replaced by an inverted-potential-vanishing-energy variant thereof. Under smoothness, convexity and coercivity hypotheses on its potential energy function, we prove, using the calculus of variations together with the Banach space implicit function theorem, the existence of a global, smooth `fundamental solution'. Higher order quantum corrections, for ground and excited states, are computed through the integration of associated systems of linear transport equations, and formal expansions for the corresponding energy eigenvalues obtained by imposing smoothness on the quantum corrections to the eigenfunctions. For linear oscillators our expansions naturally truncate, reproducing the well-known solutions for the energy eigenfunctions and eigenvalues. As an application, we calculate a number of terms in the corresponding expansions for the one-dimensional anharmonic oscillators of quartic, sectic, octic, and dectic types and find that our eigenvalue expansions agree with those of Rayleigh/Schrodinger theory, whereas our wave functions more accurately capture the more-rapid-than-gaussian decay. For the quartic oscillator our results strongly suggest that the ground state energy eigenvalue expansion and its associated wave function expansion are Borel summable to yield natural candidates for the actual exact ground state solution and its energy. Our techniques for proving the existence of the crucial `fundamental solution' to the relevant Hamilton Jacobi equation admit infinite dimensional generalizations. In a parallel project we shall show how this construction can be carried out for the Yang-Mills equations in Minkowski spacetime.