The borderline Lifshits-tail asymptotic for algebraically decaying single-impurity potentials

Let UU be a non-negative single-impurity potential in dimension dd, let N(E)N(E) denote the integrated density of states, let ρ\rho be the impurity density, and let κd\kappa_d and Cd(d+2,ρ)C_d(d+2,\rho) be the constants appearing in the Lifshits-tail asymptotics. Suppose that

limxxd+2U(x)=g\lim_{|x|\to\infty}|x|^{d+2}U(x)=g

for some constant g>0g>0.

Borderline Lifshits-tail conjecture. As E0E\downarrow0,

logN(E)[ρ2d+2(κd2E)dd+2+[Cd(d+2,ρ)]2d+2(gE)dd+2]d+22.\log N(E)\sim-\left[\rho^{\frac{2}{d+2}}\left(\frac{\kappa_d}{2E}\right)^{\frac{d}{d+2}}+\bigl[C_d(d+2,\rho)\bigr]^{\frac{2}{d+2}}\left(\frac{g}{E}\right)^{\frac{d}{d+2}}\right]^{\frac{d+2}{2}}.

The case of algebraic decay with exponent α=d+2\alpha=d+2 is identified as open because it lies at the transition between the quantum and classical Lifshits-tail regimes. The proposed formula describes the competition between the quantum kinetic contribution and the contribution from the long-range decay of the single-impurity potential.

Sources & referencesView supporting material

Primary source

Hajo Leschke, Peter Müller and Simone Warzel, “A survey of rigorous results on random Schroedinger operators for amorphous solids”, arXiv:cond-mat/0210708 (2003).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.