Abstract
We consider the existence of a dynamically stable soliton in the one-dimensional cubic-quintic nonlinear Schrödinger model with strong cubic nonlinearity management for periodic and random modulations. We show that the predictions of the averaged cubic-quintic nonlinear Schrödinger (NLS) equation and modified variational approach for the arrest of collapse coincide. The analytical results are confirmed by numerical simulations of a one-dimensional cubic-quintic NLS equation with a rapidly and strongly varying cubic nonlinearity coefficient.
Introduction: The Challenge of Wave Collapse
In many areas of physics, such as nonlinear optics and Bose-Einstein Condensates (BECs), waves are governed by the Nonlinear Schrödinger Equation (NLSE). While standard 1D systems are often stable, higher-dimensional systems or those with higher-order nonlinearities—like the "quintic" term—suffer from collapse. Collapse occurs when the attractive forces of the medium overcome the natural spreading (dispersion) of the wave, causing the pulse to shrink to a point with infinite intensity in finite time.
In a 1D cubic-quintic (CQ) model with an attractive quintic term, the "Townes soliton" is a known solution, but it is notoriously unstable: any slight perturbation causes it to either vanish or collapse. This research explores whether rapid "management" (modulation) of the cubic term can provide a dynamical "cage" to keep these solitons stable.
Methodology: Nonlinearity Management and Averaging
The authors employ three primary techniques to study the system:
- Multiscale Averaging: They consider a cubic nonlinearity coefficient $\gamma(t)$ that consists of a mean value plus a very fast, high-amplitude oscillation. By applying an averaging procedure, they derive an "averaged CQ NLSE." This new equation includes a unique term proportional to the variance of the modulation ($\sigma^2$), which acts as a stabilizing force.
- Variational Approach: To turn the complex partial differential equation (PDE) into something more manageable, they use a "chirped" ansatz (a trial function for the pulse shape). This allows them to derive ordinary differential equations (ODEs) for the pulse's width ($a$) and chirp ($b$).
- Numerical Simulations: They verify their analytical predictions using a pseudo-spectral method to solve the full 1D CQ NLSE, testing both periodic and random modulations.
Key Results: Arresting Collapse and Predictive Accuracy
The most critical insight is that the "fast" part of the nonlinearity management doesn't just average to zero. Instead, its variance ($\sigma^2$) creates a new effective potential term in the averaged equation that prevents the pulse width from reaching zero. Numerical simulations confirmed that pulses that would normally collapse in a few time units become stable for hundreds of units when management is applied.
For masses near the critical threshold ($N \approx 1.02 N_c$), the analytical formulas for the fixed point width ($a_c$) and the oscillation period ($T_c$) matched the numerical simulations with high precision. The stabilization is most effective when the pulse mass is within roughly 10% of the critical mass ($N_c$).
Discussion and Robustness
The stabilization is not limited to perfect Townes solitons. Even a Gaussian pulse, which is not a natural solution to this equation, will shed some radiation and eventually settle into a stable, oscillating soliton near the predicted fixed point. Furthermore, stabilization works for random modulations, provided a "pinning scheme" is used to ensure the accumulated nonlinearity doesn't drift over time.
While the paper mentions applications in double-doped optical fibers and BECs (using Feshbach resonance), the "strong" management required might be challenging to implement experimentally. Extending these results to 2D or 3D systems remains a significant open challenge in the field of nonlinear waves.
Summary updated