Proton radius, bound state QED and the nonlocality of the electromagnetic interaction

TL;DR

This paper identifies a fundamental gap in standard bound-state QED related to the nonlocality of electromagnetic interactions, which may explain discrepancies in proton radius measurements from hydrogen and muonic hydrogen spectroscopy.

Contribution

It introduces a new approach to bound-state QED that accounts for nonlocality and stationarity, providing corrected formulas for atomic energies and the Lamb shift.

Findings

Standard bound-state QED does not fully describe atomic states.

A new dynamical energy shift is proposed to supplement the Lamb shift.

Nonlocality of electromagnetic interaction is explicitly incorporated.

Abstract

The result of a recent measurement of the size of the proton [R. Pohl et al., Nature 466, 213] performed on the base of the muonic hydrogen spectroscopy turned out to be significantly different, by five standard deviations, from the results derived from the atomic hydrogen spectroscopy. This large discrepancy could come from the calculations of the Lamb shift in atomic hydrogen and muonic hydrogen. Here we show that there is a gap in the standard bound-state QED that may be the source of the discrepancy. This gap originates in the fact that within the framework of this theory the QED corrections are described in terms of the respective Green functions. The character of the time evolution of a system which should manifest itself in the general definition of bound states as stationary states of the system cannot be described in terms of the Green functions. We present a consistent way of solving the bound-state problem in QED starting from the condition of stationarity of the bound states. Formulae for the energies and the vectors of the states of one-electron (muon) atoms derived in this way indicate that the standard bound-state QED does not obey the exact description of the atomic states and, as a result, the Lamb shift obtained in its framework should be supplemented by an additional "dynamical" energy shift. It is shown that in this shift natural nonlocality of the electromagnetic interaction that in describing the S matrix and the Green functions is hidden in the renormalization procedure manifest itself explicitly.