Related Experiment Video
Updated: May 26, 2026

Scalable Quantum Integrated Circuits on Superconducting Two-Dimensional Electron Gas Platform
Published on: August 2, 2019
Jumps of adiabatic invariant at the separatrix of a degenerate saddle point
A V Artemyev1, A I Neishtadt, L M Zelenyi
1Space Research Institute, RAS, Profsouznaya St., 84/32, GSP-7, 117997 Moscow, Russia. artemyev@iki.rssi.ru
Abstract:
We consider a slow-fast Hamiltonian system with two degrees of freedom. One degree of freedom corresponds to slow variables, and the other one corresponds to fast variables. A characteristic ratio of the rates of change of slow and fast variables is a small parameter κ. For every fixed value of the slow variables, in the phase portrait of the fast variables there are a saddle point and separatrices passing through it. When the slow variables change, phase points may cross the separatrices. The action variable of the fast motion is an adiabatic invariant of the full system as long as a trajectory is far from the separatrices: value of the adiabatic invariant is conserved with an accuracy of order of κ on time intervals of order of 1/κ. A passage through a narrow neighborhood of the separatrices results in a jump of the adiabatic invariant. We consider a case when the saddle point is degenerate. We derive an asymptotic formula for the jump of the adiabatic invariant which turns out to be a value of order of κ(3/4) (in the case of a non-degenarate saddle point a similar jump is known to be a value of order of κ). Accumulation of these jumps after many consecutive separatrix crossings leads to the "diffusion" of the adiabatic invariant and chaotic dynamics. We verify the analytical expression for the jump of the adiabatic invariant by numerical simulations. We discuss application of the obtained results to the description of charged particle dynamics in the Earth magnetosphere.
Related Concept Videos
Adiabatic Processes for an Ideal Gas
Separable Differential Equations
Entropy Change in Reversible Processes
The statement can be further generalized to prove that entropy is a state function. Take a cyclic process between any two points on a p-V diagram.
Path Between Thermodynamics States
The Entropy as a State Function
Work Done in an Adiabatic Process

