This talk concerns the (generalized) Soler model: a nonlinear (massive) Dirac equation with a nonlinearity taking the form of a space-dependent mass. The equation admits standing wave solutions and they are generally expected to be stable (i.e., small perturbations in the initial conditions stay small) based on numerical simulations. However, contrarily to the nonlinear Schrödinger equation for example, there are very few results in this direction. The results that I will discuss concern the simpler question of spectral stability (and instability), i.e., the absence (or presence) of exponentially growing solutions to the linearized equation around a solitary wave. As in the case of the nonlinear Schrödinger equation, this is equivalent to the presence or absence of « unstable eigenvalues » of a non-self-adjoint operator with a particular block structure. I will highlight the differences and similarities with the Schrödinger case, present some partial results for the one-dimensional case, and discuss open problems.

This is joint work with Danko Aldunate, Edgardo Stockmeyer and Hanne van den Bosch.
https://indico.math.cnrs.fr/event/7322/