Theoretical and experimental investigation of Sona transitions

In 1967, P. G. Sona proposed a method to increase nuclear polarization in polarized hydrogen and deuterium ion sources. The technique uses a static magnetic field produced by two opposing solenoid coils. Applying a semi-classical approach, he designed the magnetic field distribution of the device. It consists of a longitudinal (parallel to the beam axis) and a transverse component with axial symmetry. One of the requirements for an efficient performance is a non-adiabatic passage through the low field region. This leads to the zero-crossing requirements, which render Sona transitions inapplicable to thermal sources. For more than 50 years, Sona transitions have been implemented to several experiments involving various types of ground-state and metastable sources (e.g. H, D, ${}^3$He, etc.). However, an unanticipated behavior of polarization was observed in two of them. The present study provides experimental and theoretical insights into this effect. For this purpose, an experimental setup, consisting of components of a Lamb-shift polarimeter, has been developed. This can be applied to metastable hydrogen and deuterium sources. The recorded signals reveal that there is an interaction between the hyperfine substates of the atoms, which results in transitions between them. The transitions cannot be described by Sona's work and they are observed even while the zero-crossing conditions are fulfilled. Thus, a purely quantum mechanical treatment is employed to explore the spin dynamics in a Sona transition unit. In particular, the theoretical analysis is realized by means of time-dependent perturbation theory applied in the rest frame of the atoms. The system of coupled first order differential equations for the interplay between the states is derived analytically in the hyperfine regime. Its solution is solely feasible with numerical methods. The dependence of the results on the magnetic field configuration and the beam characteristics (kinetic energy, size, position, etc.) is also examined. The simulated and recorded signals are compared and show a very good correlation. Small deviations are noticed, but they are mainly caused by the experimental instabilities. Upgrades to the production of polarized particles are in progress with the intent of improving beam intensity and stability. A new Sona field configuration will be developed with a sinusoidal distribution in order to simplify the study of spin dynamics and the field parameterization in the simulation program. These two improvements, along with the development of data analysis code, can lead to performing precision measurements with such an experimental setup. Eventually, the proposed concept could serve as a spectroscopic tool for testing QED corrections in H or D atoms and more complex atomic/molecular/ionic systems.