Attractive and driven interactions in quantum dots: Mechanisms for geometric pumping

Artikel i vetenskaplig tidskrift, 2018

We analyze time-dependent transport through a quantum dot with electron-electron interaction that is statically tunable to both repulsive and attractive regimes, or even dynamically driven. Motivated by the recent experimental realization [A. Hamo, Nature (London) 535, 395 (2016)NATUAS0028-083610.1038/nature18639] of such a system in a static double quantum dot we compute the geometric pumping of charge in the limit of weak tunneling, high temperature, and slow driving. We analyze the responses for all possible pumping experiments or "driving protocols", each defined by choosing a pair of driving parameters (gate voltage, bias voltage, tunnel coupling, electron-electron interaction). We show that such responses for different experiments can be governed by a common, underlying pumping mechanism, which is characterized by a set of effective parameters. The latter are nontrivial combinations of the experimentally driven parameters and other static parameters. If two different pumping experiments result in the same modulation of the effective parameters, i.e., the underlying mechanism is the same, then their responses will also be the same. Interestingly, for static attractive interaction we find a nonzero pumping response despite the attractive Coulomb blockade that hinders stationary transport. Furthermore, we identify a unique pumping response whose underlying mechanism relies on the interaction to be one of the driving parameters: it cannot be obtained with other sets of driving parameters. Finally, although a single-dot model with orbital pseudospin describes most of the physics of the mentioned experimental setup, it is crucial to account for the additional (real-)spin degeneracy of the double dot and the associated electron-hole symmetry breaking. This is necessary because the pumping response is more sensitive than dc transport measurements and detects this difference through pronounced qualitative effects.