Low-energy effective theory of localization-delocalization transition in noninteger-charged electron wave packets

Abstract

We present a low-energy effective theory to describe the localization-delocalization transition, which occurs for wave functions of electrons and holes injected individually by a voltage pulse with noninteger flux quantum. We find that the transition can be described by an effective scattering matrix in a truncated low-energy space, which is composed of two parts. The first part describes the infrared-divergence of the scattering matrix, while the second part represents the high-energy correlation. For short-tailed pulses which decay faster than Lorentzian, the scattering matrix exhibits solely an inverse linear divergence in the infrared limit. The divergence is responsible for the dynamical orthogonality catastrophe, which leads to electron-hole pairs with delocalized wave functions. In contrast, the high-energy correlation can be approximated by a constant term, which leads to electron-hole pairs with localized wave functions. Due to the competition between the two terms, the wave functions can undergo a localization-delocalization transition, which occurs for electrons and holes injected individually by the voltage pulse. As a consequence, the localization-delocalization transitions for all short-tailed pulses can be described by the same effective scattering matrix, suggesting that they belong to the same universality class. For pulses with longer tails, the scattering matrix can exhibit additional infrared-divergences. We show that a Lorentzian pulse gives rise to a logarithmic divergence, while a fractional-powered Lorentzian pulse gives rise to a power-law divergence. The additional divergence can lead to localization-delocalization transitions belonging to different universality classes. These results demonstrate the fine-tuning capabilities of the localization-delocalization transition in time-dependent quantum transport.