Leschke's research publications listed below all refer to properties of non-relativistic quantum systems which are modeled by some
Hamiltonian, that is, by some
self-adjoint operator on
Hilbert space representing the total energy of the system, possibly depending on
random variables simulating disorder. In the publications from 2000 to 2017 the Hamiltonian is of
Schrödinger type, that is, an operator for the sum of the kinetic and potential energy of "point-like" particles in
Euclidean space. The two publications with Kurt Broderix (1962–2000) extend previously known continuity properties of the corresponding one-parameter Schrödinger
semi-group (or Gibbs operator for different temperatures) to rather general magnetic fields and to (random) potential fields possibly leading to unbounded semi-groups; by suitably extending the
Feynman–Kac formula and using the
diamagnetic inequality. The other three publications from 2000 to 2004 consider the case of a single particle subject to a constant magnetic field and a
random potential field. For a Poissonian field with positive single-impurity potential U the low-energy behavior of the integrated (or cumulative)
density of states is derived, depending on the range of U. For a Gaussian random field (without an underlying lattice structure) the first proofs are given for the existence of the density of states and of
Anderson localization in multi-dimensional continuous space. The publications in 2014 and 2017 refer to the case of many non-interacting particles which obey
Fermi–Dirac statistics. For the corresponding ideal
Fermi gas in
thermodynamic equilibrium they contain the first rigorous results on the asymptotic growth of its quantum Rényi
entropies of (spatial) entanglement at arbitrary temperature. These results have served as a standard of comparison for approximate arguments and/or numerical methods to better understand the correlations in many-fermion systems with interaction. The publications in 2021 are among the first ones providing rigorous results on quantum versions of the classic(al)
Sherrington–Kirkpatrick spin-glass model. In particular, they prove for the first time the existence of a phase transition (related to
spontaneous replica-symmetry breaking) if the temperature and the strength of the "transverse" magnetic field are low enough. The publication in 2023 illuminates this phase transition's relevance to the
quantum-annealing algorithm in computer science. == Selected publications since 2000 ==