We propose a method for observation of the quasi-stationary states of neutrons, localized near the curved mirror surface. The bounding effective well is formed by the centrifugal potential and the mirror Fermi-potential. This phenomenon is an example
of an exactly solvable quantum bouncer problem that could be studied experimentally. It could provide a promising tool for studying fundamental neutron-matter interactions, as well as quantum neutron optics and surface physics effects. We develop formalism, which describes quantitatively the neutron motion near the mirror surface. The effects of mirror roughness are taken into account.
The available data on neutron scattering were analyzed to constrain a hypothetical new short-range interaction. We show that these constraints are several orders of magnitude better than those usually cited in the range between 1 pm and 5 nm. This di
stance range occupies an intermediate space between collider searches for strongly coupled heavy bosons and searches for new weak macroscopic forces. We emphasise the reliability of the neutron constraints in so far as they provide several independent strategies. We have identified the most promising way to improve them.
