OctoInformBeta

Localization and delocalization in the Anderson model

Fix h>0h>0, and let v=(vx)x∈ℤ2v=(v_x)_{x\in\mathbb Z^2} have independent coordinates uniformly distributed on [−h,h][-h,h]. The square-lattice Anderson operator on ℓ2(ℤ2)\ell^2(\mathbb Z^2) is defined by (Hvu)(x)=∑|y−x|1=1u(y)+vxu(x)(H_vu)(x)=\sum_{|y-x|_1=1}u(y)+v_xu(x). Pure-point spectral type means that the closed span of the eigenvectors is the whole Hilbert space. The planar spectral localization question asks whether HvH_v almost surely has pure-point spectral type throughout its spectrum for every fixed positive disorder strength.

The preprint proves this pure-point assertion for the uniform single-site law. The preprint also derives almost-sure density of the eigenvalues in [−4−h,4+h][-4-h,4+h].

For every fixed h>0h>0, the Lean library PlanarAnderson proves the almost-sure existence of a self-adjoint Anderson operator with spectrum [−4−h,4+h][-4-h,4+h]. The Lean library’s conclusion of pure-point spectral type requires a complete orthonormal eigenbasis as a hypothesis. The Lean library does not prove the existence of such a basis for the random operator. The Lean library does not prove unconditional density of its eigenvalues. Lean has checked every result of the Lean library, so the Lean library is the reference for the statements on these pages. Each result is stated as the Lean library states it and linked to the Lean declaration behind it. Where the preprint gives a different statement, its statement appears beside the Lean library’s statement. The preprint serves only as a guide to the order and names of the results.