Inverting a normal harmonic oscillator: Physical interpretation and applications
Abstract
A harmonic oscillator with time-dependent mass m(t) and a time-dependent (squared) frequency ω2(t) occurs in the modelling of several physical systems. It is generally believed that systems, with m(t)>0 and ω2(t)>0 (normal oscillator) are stable while systems with m(t)>0 and ω2(t)<0 (inverted oscillator) are unstable. We show that it is possible to represent the same physical system either as a normal oscillator or as an inverted oscillator by redefinition of dynamical variables. While we expect the physics to be invariant under such redefinitions, it is not obvious how this invariance actually comes about. We study the relation between these two, normal and inverted, representations of an oscillator in detail both in Heisenberg and Schr\"odinger pictures to clarify several conceptual and technical issues. The situation becomes more involved when the oscillator is coupled to another (semi)classical degree of freedom C(t) and we want to study the back-reaction of the quantum system q(t) on C(t), in the semi-classical approximation. We provide a simple prescription for the back-reaction based on energy conservation and study that the dynamics of the full system in both normal and the inverted oscillator representation. The physics again remains invariant but there are some extra subtleties which we clarify. The implications of these results for quantum field theory in cosmological backgrounds are discussed briefly in an appendix.
Turn this paper into a lesson
ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.