Abstract:We present numerical minimizers for the first seven eigenvalues of the magnetic Dirichlet Laplacian with constant magnetic field in a wide range of field strengths. Adapting an approach by Antunes and Freitas, we use gradient descent for the minimization procedure together with the Method of Fundamental solutions for eigenvalue computation. Remarkably, we observe that when the magnetic flux exceeds the index of the target eigenvalue, the minimizer is always a disk.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is, given the magnetic field strength \(B\), to find the geometric shape that minimizes the first seven eigenvalues of the magnetic Dirichlet Laplacian operator. Specifically, the author is concerned with how the solutions to these eigenvalue minimization problems change under different magnetic field strengths, and particularly notes that when the magnetic flux exceeds the index of the target eigenvalue, the minimizing solution is always a circular region.
### Problem Background
The eigenvalue problem of the magnetic Dirichlet Laplacian operator with a constant magnetic field \(B\) on a two - dimensional bounded open set \(\Omega\subset\mathbb{R}^2\) is considered in the research:
\[
(-i\nabla + A)^2u=\lambda u,\quad\text{in }\Omega,
\]
\[
u = 0,\quad\text{on }\partial\Omega,
\]
where the vector potential \(A(x_1,x_2)=\frac{B}{2}\begin{pmatrix}-x_2\\x_1\end{pmatrix}\) is the standard linear vector potential.
From a physical perspective, this problem describes the energy levels of a charged quantum - mechanical particle confined within \(\Omega\) and interacting with a constant magnetic field perpendicular to the plane.
### Research Motivation and Problem Statement
For a general geometric shape \(\Omega\), the eigenvalues of the magnetic Dirichlet Laplacian operator \(H_{\Omega}^B\) represent the energy levels that the particle is allowed to have. The goal of the paper is to find, by numerical methods, the geometric shape that minimizes the first seven eigenvalues and to study how these minimizing solutions change with the magnetic field strength \(B\).
### Main Findings
- **Eigenvalue Minimization**: The author uses the gradient descent method and the Method of Fundamental Solutions (MFS) to calculate the eigenvalues and perform minimization.
- **Disk Optimality**: A remarkable observation is that when the magnetic flux exceeds the index of the target eigenvalue, the minimizing solution is always a circular region. For example, for the first eigenvalue \(\lambda_1\), Erdős proved in 1996 that the circular region is optimal (generalizing the Faber - Krahn inequality).
- **Higher - Order Eigenvalues**: For higher - order eigenvalues, the author proposes new conjectures, especially regarding the behavior of the minimizing solutions under strong magnetic fields.
### Numerical Methods
To solve this optimization problem, the author adopts the following methods:
1. **Method of Fundamental Solutions (MFS)**: Used to calculate the eigenvalues of the magnetic Dirichlet Laplacian operator.
2. **Gradient Descent Method**: Used to optimize the geometric shape to minimize the eigenvalues.
### Conclusion
Through numerical experiments, the author not only verifies existing theoretical results (such as Erdős' theorem), but also proposes new conjectures for the minimization problem of higher - order eigenvalues. In particular, they find that when the magnetic flux exceeds the eigenvalue index, the minimizing solution is always a circular region, which provides valuable clues for further research.