Low-temperature electron dephasing time in AuPd revisited

Abstract

Ever since the first discoveries of the quantum-interference transport in mesoscopic systems, the electron dephasing times, $\tau_\phi$, in the concentrated AuPd alloys have been extensively measured. The samples were made from different sources with different compositions, prepared by different deposition methods, and various geometries (1D narrow wires, 2D thin films, and 3D thickfilms) were studied. Surprisingly, the low-temperature behavior of $\tau_\phi$ inferred by different groups over two decades reveals a systematic correlation with the level of disorder of the sample. At low temperatures, where $\tau_\phi$ is (nearly) independent of temperature, a scaling $\tau_\phi^{\rm max} \propto D^{-\alpha}$ is found, where $tau_\phi^{\rm max}$ is the maximum value of $\tau_\phi$ measured in the experiment, $D$ is the electron diffusion constant, and the exponent $\alpha$ is close to or slightly larger than 1. We address this nontrivial scaling behavior and suggest that the most possible origin for this unusual dephasing is due to dynamical structure defects, while other theoretical explanations may not be totally ruled out.Comment: to appear in Physica E, Proceedings for the International Seminar and Workshop "Quantum Coherence, Noise, and Decoherence in Nanostructures", 15-26 May 2006, Dresde

Weak localization, Aharonov-Bohm oscillations, and decoherence in arrays of quantum dots

“…E.g., in order to be able to attribute dephasing times as short as τ ϕ0 10 −12 s to magnetic impurities one needs to assume huge concentration of such impurities ranging from few hundreds to few thousands ppm which appears highly unrealistic, in particular for systems like carbon nanotubes, 2DEGs or quantum dots. Similar arguments were independently put forward by Lin and coworkers 47,49 .…”

“…E.g., in order to be able to attribute dephasing times as short as τ ϕ0 10 −12 s to magnetic impurities one needs to assume huge concentration of such impurities ranging from few hundreds to few thousands ppm which appears highly unrealistic, in particular for systems like carbon nanotubes, 2DEGs or quantum dots. Similar arguments were independently put forward by Lin and coworkers 47,49 . Thus, although electron dephasing due to scattering on magnetic impurities is by itself an interesting issue, its role in low temperature saturation of τ ϕ in disordered conductors is sometimes strongly overemphasized.…”

“…As our theory of dephasing by electron-electron interactions predicts a rather steep increase of τ ϕ0 with the system diffusion coefficient D, for most weakly disordered metals as τ ϕ0 ∝ D 3 , we can conclude that for a large number of disordered conductors τ ϕ0 strongly increases with increasing D. This trend is indeed quite obvious for relatively weakly disordered conductors. On the other hand, Lin and coworkers 9,[47][48][49] analyzed numerous experimental data for τ ϕ0 obtained by various groups in rather strongly disordered conductors with D 10 cm 2 /s and observed systematic decrease of τ ϕ0 with increasing D. The data could be fitted by the dependence τ ϕ0 ∝ D −α with the power α 1. This trend is clearly just the opposite to one observed in less disordered conductors with D 10 cm 2 /s.…”

The role of exceptional points in quantum systems

“…According [81], one consensus has been reached by several groups, saying that the responsible electron dephasing processes in highly disordered and weakly disordered metals might be dissimilar. That means, while one mechanism is responsible for dephasing in weakly disordered metals, another mechanism may be relevant for the saturation (or very weak temperature dependence) of τ φ found in highly disordered alloys.…”

“…That means, while one mechanism is responsible for dephasing in weakly disordered metals, another mechanism may be relevant for the saturation (or very weak temperature dependence) of τ φ found in highly disordered alloys. According to the authors of [81], the intriguing electron dephasing is very unlikely due to magnetic scattering. It may originate from specific dynamical structure defects in the samples.…”