Hyperfine frequency shift in two-dimensional atomic hydrogen

TL;DR

This paper explains the unexpectedly small hyperfine frequency shift in 2D atomic hydrogen on superfluid helium by considering density-dependent wall and pressure shifts, revealing new effects specific to 2D hydrogen and correcting previous scattering length measurements.

Contribution

It introduces the concept of density-dependent wall shift in 2D hydrogen and provides a corrected value for the difference in scattering lengths, advancing understanding of hyperfine interactions in 2D atomic systems.

Findings

The hyperfine shift is mainly due to density-dependent wall and pressure shifts.

The density-dependent wall shift is a novel concept first mentioned in this work.

The difference in triplet and singlet scattering lengths is half of previous estimates.

Abstract

We propose the explanation of a surprisingly small hyperfine frequency shift in the two-dimensional (2D) atomic hydrogen bound to the surface of superfluid helium below 0.1 K. Owing to the symmetry considerations, the microwave-induced triplet-singlet transitions of atomic pairs in the fully spin-polarized sample are forbidden. The apparent nonzero shift is associated with the density-dependent wall shift of the hyperfine constant and the pressure shift due to the presence of H atoms in the hyperfine state $a$ not involved in the observed $b\to c$ transition. The interaction of adsorbed atoms with one another effectively decreases the binding energy and, consequently, the wall shift by the amount proportional to their density. The pressure shift of the $b\to c$ resonance comes from the fact that the impurity $a$-state atoms interact differently with the initial $b$-state and final $c$-state atoms and is also linear in density. The net effect of the two contributions, both specific for 2D hydrogen, is comparable with the experimental observation. To our knowledge, this is the first mentioning of the density-dependent wall shift. We also show that the difference between the triplet and singlet scattering lengths of H atoms, $a_t-a_s=30(5)$ pm, is exactly twice smaller than the value reported by Ahokas {\it et al.}, Phys. Rev. Lett. {\bf101}, 263003 (2008).